Inflation is most often described using quantum field theory (QFT) on a fixed, curved spacetime background. Such a description is valid only if the spatial volume of the region considered is so large that its size and shape moduli behave classically. However, if we trace an inflating universe back to early times, the volume of any comoving region of interest -- for example the present Hubble volume -- becomes exponentially small. Hence, quantum fluctuations in the trajectory of the background cannot be neglected at early times. In this paper, we develop a path integral description of a flat, inflating patch (approximated as de Sitter spacetime), treating both the background scale factor and the gravitational wave perturbations quantum mechanically. We find this description fails at small values of the initial scale factor, because \emph{two} background saddle point solutions contribute to the path integral. This leads to a breakdown of QFT in curved spacetime, causing the fluctuations to be unstable and out of control. We show the problem may be alleviated by a careful choice of quantum initial conditions, for the background and the fluctuations, provided that the volume of the initial, inflating patch is larger than $\gg H^{-1}$ in Planck units with $H$ the Hubble constant at the start of inflation. The price of the remedy is high: not only the inflating background, but also the stable, Bunch-Davies fluctuations must be input by hand. Our discussion emphasizes that, even if the inflationary scale is far below the Planck mass, new physics is required to explain the initial quantum state of the universe.

There is increasing evidence from string theory that effective field theories are only reliable over approximately sub-Planckian field excursions. The two most promising effective models for early universe cosmology, inflation and ekpyrosis, are mechanisms that, in order to address cosmological puzzles, must operate over vast expansion/energy ranges. This suggests that it might be appropriate to describe them using scaling laws. Here we combine these two ideas and demonstrate that they drive inflation and ekpyrosis to their extremes: inflation must start at ultra-slow-roll, and ekpyrosis at ultra-fast-roll. At face value, the implied spectra are overly tilted to the red, although in both cases minor departures from pure scale freedom bring the spectral indices within current observational bounds. These models predict a significant spectral running at a level detectable in the near future ($\alpha_s \approx -10^{-3}$). Ekpyrotic models with minimal coupling are nevertheless ruled out, as they lead to levels of non-Gaussianity that are at least an order of magnitude too large. No such restrictions apply to models with a kinetic coupling between the two ekpyrotic scalar fields, and these remain the most promising ekpyrotic models. An additional swampland criterion that was recently proposed for the slope of the scalar field potential would however rule out all ultra-slow-roll models of inflation. Finally, we speculate on the existence of corresponding restrictions on the slope at negative potential values, which might lead to similarly severe constraints on ekpyrotic models.

When coupling fermions to gravity, torsion is naturally induced. We consider the possibility that fermion bilinears can act as a source for torsion, altering the dynamics of the early universe such that the big bang gets replaced with a classical non-singular bounce. We extend previous studies in several ways: we allow more general fermion couplings, consider both commuting and anti-commuting spinors, and demonstrate that with an appropriate choice of potential one can easily obtain essentially arbitrary equations of state, including violations of the null energy condition, as required for a bounce. As an example, we construct a model of ekpyrotic contraction followed by a non-singular bounce into an expanding phase. We analyze cosmological fluctuations in these models, and show that the perturbations can be rewritten in real fluid form. We find indications that spinor bounces are stable, and exhibit several solutions for the perturbations. Interestingly, spinor models do not admit a scalar-vector-tensor decomposition, and consequently some types of scalar fluctuations can act as a source for gravitational waves already at linear order. We also find that the first order dynamics are directionally dependent, an effect which might lead to distinguished observational signatures.

We study two-field bouncing cosmologies in which primordial perturbations are created in either an ekpyrotic or a matter-dominated contraction phase. We use a non-singular ghost condensate bounce model to follow the perturbations through the bounce into the expanding phase of the universe. In contrast to the adiabatic perturbations, which on large scales are conserved across the bounce, entropy perturbations can grow significantly during the bounce phase. If they are converted into adiabatic/curvature perturbations after the bounce, they typically form the dominant contribution to the observed temperature fluctuations in the microwave background, which can have several beneficial implications. For ekpyrotic models, this mechanism loosens the constraints on the amplitude of the ekpyrotic potential while naturally suppressing the intrinsic amount of non-Gaussianity. For matter bounce models, the mechanism amplifies the scalar perturbations compared to the associated primordial gravitational waves.

In the framework of scalar-tensor theories of gravity, we construct a new kind of cosmological model that conflates inflation and ekpyrosis. During a phase of conflation, the universe undergoes accelerated expansion, but with crucial differences compared to ordinary inflation. In particular, the potential energy is negative, which is of interest for supergravity and string theory where both negative potentials and the required scalar-tensor couplings are rather natural. A distinguishing feature of the model is that, for a large parameter range, it does not significantly amplify adiabatic scalar and tensor fluctuations, and in particular does not lead to eternal inflation and the associated infinities. We also show how density fluctuations in accord with current observations may be generated by adding a second scalar field to the model. Conflation may be viewed as complementary to the recently proposed anamorphic universe of Ijjas and Steinhardt.

We explicitly confirm that spatially flat non-singular bouncing cosmologies make sense as effective theories. The presence of a non-singular bounce in a spatially flat universe implies a temporary violation of the null energy condition, which can be achieved through a phase of ghost condensation. We calculate the scale of strong coupling and demonstrate that the ghost-condensate bounce remains trustworthy throughout, and that all perturbation modes within the regime of validity of the effective description remain under control. For this purpose we require the perturbed action up to third order in perturbations, which we calculate in both flat and co-moving gauge -- since these two gauges allow us to highlight different physical aspects. Our conclusion is that there exist healthy descriptions of non-singular bouncing cosmologies providing a viable resolution of the big-bang singularities in cosmological models. Our results also suggest a variant of ekpyrotic cosmology, in which entropy perturbations are generated during the contracting phase, but are only converted into curvature perturbations after the bounce.

Employing the covariant formalism, we derive the evolution equations for two scalar fields with non-canonical field space metric up to third order in perturbation theory. These equations can be used to derive predictions for local bi- and trispectra of multi-field cosmological models. Our main application is to ekpyrotic models in which the primordial curvature perturbations are generated via the non-minimal entropic mechanism. In these models, nearly scale-invariant entropy perturbations are generated first due to a non-minimal kinetic coupling between two scalar fields, and subsequently these perturbations are converted into curvature perturbations. Remarkably, the entropy perturbations have vanishing bi- and trispectra during the ekpyrotic phase. However, as we show, the conversion process to curvature perturbations induces local non-Gaussianity parameters $f_{NL}$ and $g_{NL}$ at levels that should be detectable by near-future observations. In fact, in order to obtain a large enough amplitude and small enough bispectrum of the curvature perturbations, as seen in current measurements, the conversion process must be very efficient. Interestingly, for such efficient conversions the trispectrum parameter $g_{NL}$ remains negative and typically of a magnitude ${\cal O}(10^2) - {\cal O}(10^3),$ resulting in a distinguishing feature of non-minimally coupled ekpyrotic models.

We calculate the running of the scalar index in the ekpyrotic and matter bounce cosmological scenarios, and find that it is typically negative for ekpyrotic models, while it is typically positive for realizations of the matter bounce where multiple fields are present. This can be compared to inflation, where the observationally preferred models typically predict a negative running. The magnitude of the running is expected to be between $10^{-4}$ and up to $10^{-2},$ leading in some cases to interesting expectations for near-future observations.

This paper investigates the manner in which classical universes are obtained in the no-boundary quantum state. In this context, universes can be characterised as classical (in a WKB sense) when the wavefunction is highly oscillatory, i.e. when the ratio of the change in the amplitude of the wavefunction becomes small compared to the change in the phase. In the presence of a positive or negative exponential potential, the WKB condition is satisfied in proportion to a factor $e^{-(\epsilon - 3)N/(\epsilon -1)},$ where $\epsilon$ is the (constant) slow-roll/fast-roll parameter and $N$ designates the number of e-folds. Thus classicality is reached exponentially fast in $N$, but only when $\epsilon < 1$ (inflation) or $\epsilon > 3$ (ekpyrosis). Furthermore, when the potential switches off and the ekpyrotic phase goes over into a phase of kinetic domination, the level of classicality obtained up to that point is preserved. Similar results are obtained in a cyclic potential, where a dark energy plateau is added. Finally, for a potential of the form $-\phi^n$ (with $n=4,6,8$), where the classical solution becomes increasingly kinetic-dominated, there is an initial burst of classicalisation which then quickly levels off. These results demonstrate that inflation and ekpyrosis, which are the only dynamical mechanisms known for smoothing the universe, share the perhaps even more fundamental property of rendering space and time classical in the first place.

Ekpyrotic instantons describe the emergence of classical contracting universes out of the no-boundary quantum state. However, up to now these instantons ended in a big crunch singularity. We remedy this by adding a higher-derivative term, allowing a ghost condensate to form. This causes a smooth, non-singular bounce from the contracting phase into an expanding, kinetic-dominated phase. Remarkably, and although there is a non-trivial evolution during the bounce, the wavefunction of the universe is "classical" in a WKB sense just as much after the bounce as before. These new non-singular instantons can thus form the basis for a fully non-singular and calculable ekpyrotic history of the universe, from creation until now.

The no-boundary proposal provides a compelling theory for the initial conditions of our universe. We study the implications of such initial conditions for ekpyrotic and cyclic cosmologies. These cosmologies allow for the existence of a new type of "ekpyrotic instanton", which describes the creation of a universe in the ekpyrotic contraction phase. Remarkably, we find that the ekpyrotic attractor can explain how the universe became classical. In a cyclic context, in addition to the ekpyrotic instantons there exist de Sitter-like instantons describing the emergence of the universe in the dark energy phase. Our results show that typically the ekpyrotic instantons yield a higher probability. In fact, in a potential energy landscape allowing both inflationary and cyclic cosmologies, the no-boundary proposal implies that the probability for ekpyrotic and cyclic initial conditions is vastly higher than that for inflationary ones.

We present a new class of complex instantons in the context of ekpyrotic cosmological theories. These instantons, which satisfy the "no-boundary" boundary conditions, describe the emergence of a classical, contracting universe out of nothing. The ekpyrotic attractor is essential in guaranteeing an evolution towards a real, Lorentzian history of the universe. In the context of the no-boundary proposal, the relative probability for such ekpyrotic histories compared to inflationary ones is very high -- in fact, assuming a bounce can be incorporated, these new instantons currently describe the most likely origin of the universe.

We explore a new type of entropic mechanism for generating density perturbations in a contracting phase in which there are two scalar fields, but only one has a steep negative potential. This first field dominates the energy density and is the source of the ekpyrotic equation of state. The second field has a negligible potential, but its kinetic energy density is coupled to the first field with a non-linear sigma-model type interaction. We show that for any ekpyrotic equation of state it is possible to choose the potential and the kinetic coupling such that exactly scale-invariant (or nearly scale-invariant) entropy perturbations are produced. The corresponding background solutions are stable, and the bispectrum of the entropy perturbations vanishes as no non-Gaussianity is produced during the ekpyrotic phase. Hence, the only contribution to non-Gaussianity comes from the non-linearity of the conversion process during which entropic perturbations are turned into adiabatic ones, resulting in a local non-Gaussianity parameter $f_{NL} \sim 5$.

We study the propagation of super-horizon cosmological perturbations in a non-singular bounce spacetime. The model we consider combines a ghost condensate with a Galileon term in order to induce a ghost-free bounce. Our calculation is performed in harmonic gauge, which ensures that the linearized equations of motion remain well-defined and non-singular throughout. We find that, despite the fact that near the bounce the speed of sound becomes imaginary, super-horizon curvature perturbations remain essentially constant across the bounce. In fact, we show that there is a time close to the bounce where curvature perturbations of all wavelengths are required to be momentarily exactly constant. We relate our calculations to those performed in other gauges, and comment on the relation to previous results in the literature.

We investigate the effect of non-canonical kinetic terms on inflation in supergravity. We find that the biggest impact of such higher-derivative kinetic terms is due to the corrections to the potential that they induce via their effect on the auxiliary fields, which now have a cubic equation of motion. This is in contrast to the usual (non-supersymmetric) effective field theory expansion which assumes that mass-suppressed higher-derivative terms do not affect the lower-derivative terms already present. We demonstrate with several examples that such changes in the potential can significantly modify the inflationary dynamics. Our results have immediate implications for effective descriptions of inflation derived from string theory, where higher-derivative kinetic terms are generally present. In addition we elucidate the structure of the theory in the parameter range where there are three real solutions to the auxiliary field's equation of motion, studying the resulting three branches of the theory, and finding that one of them suffers from a singularity in the speed of propagation of fluctuations.

The entropic mechanism for producing nearly scale-invariant density perturbations in a contracting ekpyrotic universe relies on having an unstable scalar potential. Here we develop a variant of this mechanism (recently proposed by Qiu, Gao and Saridakis, and by Li), in which there exists a non-trivial coupling between adiabatic and entropic fields, and where an unstable potential is not required. In the model nearly scale-invariant entropy perturbations are generated first. Remarkably, we find that the bispectrum of these perturbations vanishes, with the values of the non-Gaussianity parameters of local, equilateral and orthogonal type all exactly zero. Subsequently, the entropy perturbations can be converted into curvature perturbations by a variety of mechanisms. The bispectrum of the curvature perturbations depends on the non-linearity of the conversion process and is thus more model-dependent - however, for an efficient conversion process the final bispectrum remains small. The only distinguishing feature compared to single-field slow-roll inflationary models is an absence of primordial gravitational waves. Thus the present model provides a perfect match to current data from the PLANCK satellite.

We examine the processes of quantum squeezing and decoherence of density perturbations produced during a slowly contracting ekpyrotic phase in which entropic perturbations are converted to curvature perturbations before the bounce to an expanding phase. During the generation phase, the entropic fluctuations evolve into a highly squeezed quantum state, analogous to the evolution of inflationary perturbations. Subsequently, during the conversion phase, quantum coherence is lost very efficiently due to the interactions of entropy and adiabatic modes. Moreover, while decoherence occurs, the adiabatic curvature perturbations inherit their semi-classicality from the entropic perturbations. Our results confirm that, just as for inflation, an ekpyrotic phase can generate nearly scale-invariant curvature perturbations which may be treated as a statistical ensemble of classical density perturbations, in agreement with observations of the cosmic background radiation.

We show that results from the Planck satellite reported in 2013 are consistent with the simplest cyclic models for natural parameter ranges i.e., order unity dimensionless coefficients, assuming the standard entropic mechanism for generating curvature perturbations. With improved precision, forthcoming results from Planck and other experiments should be able to test the parameter ranges by confirming or refuting the core predictions - i.e., no observable primordial B-mode polarization and detectable local non-gaussianity. A new prediction, given the Planck 2013 constraints on the bispectrum, is a sharp constraint on the local trispectrum parameter g_{NL}; namely, the simplest models predict it is negative, with g_{NL} < -1700.

Galileons are higher-derivative theories of a real scalar which nevertheless admit second order equations of motion. They have interesting applications as dark energy models and in early universe cosmology, and have been conjectured to arise as descriptions of brane dynamics in string theory. In the present paper, we study the bosonic sector of globally N=1 supersymmetric extensions of the cubic Galileon Lagrangian in detail. Supersymmetry requires that the Galileon scalar now becomes paired with a second real scalar field. We prove that the presence of this second scalar causes the equations of motion to become higher than second order, thus leading to the appearance of ghosts. We also analyze the energy scales up to which, in an effective field theory description, the ghosts can be tamed.

We present the theory of a supersymmetric ghost condensate coupled to N=1 supergravity. This is accomplished using a general formalism for constructing locally supersymmetric higher-derivative chiral superfield actions. The theory admits a ghost condensate vacuum in de Sitter spacetime. Expanded around this vacuum, the scalar sector of the theory is shown to be ghost-free with no spatial gradient instabilities. By direct calculation, the fermion sector is found to consist of a massless chiral fermion and a massless gravitino. By analyzing the supersymmetry transformations, we find that the chiral fermion transforms inhomogeneously, indicating that the ghost condensate vacuum spontaneously breaks local supersymmetry with this field as the Goldstone fermion. Although potentially able to get a mass through the super-Higgs effect, the vanishing superpotential in the ghost condensate theory renders the gravitino massless. Thus local supersymmetry is broken without the super-Higgs effect taking place. This is in agreement with, and gives an explanation for, the direct calculation.

It was recently demonstrated that, when coupled to N=1 supergravity, the Dirac-Born-Infeld (DBI) action constructed from a single chiral superfield has the property that when the higher-derivative terms become important, the potential becomes negative. Thus, DBI inflation cannot occur in its most interesting, relativistic regime. In this paper, it is shown how to overcome this problem by coupling the model to one or more additional chiral supermultiplets. In this way, one obtains effective single real scalar field DBI models with arbitrary positive potentials, as well as multiple real scalar field DBI inflation models with hybrid potentials.

We construct N=1 supergravity extensions of scalar field theories with higher-derivative kinetic terms. Special attention is paid to the auxiliary fields, whose elimination leads not only to corrections to the kinetic terms, but to new expressions for the potential energy as well. For example, a potential energy can be generated even in the absence of a superpotential. Our formalism allows one to write a supergravity extension of any higher-derivative scalar field theory and, therefore, has applications to both particle physics and cosmological model building. As an illustration, we couple the higher-derivative DBI action describing a 3-brane in 6-dimensions to N=1 supergravity. This displays a number of new features-- including the fact that, in the regime where the higher-derivative kinetic terms become important, the potential tends to be everywhere negative.

Eternal inflation produces pocket universes with all physically allowed vacua and histories. Some of these pocket universes might contain a phase of slow-roll inflation, some might undergo cycles of cosmological evolution and some might look like the galilean genesis or other "emergent" universe scenarios. Which one of these types of universe we are most likely to inhabit depends on the measure we choose in order to regulate the infinities inherent in eternal inflation. We show that the currently leading measure proposals, namely the global light-cone cut-off and its local counterpart, the causal diamond measure, as well as closely related proposals, all predict that we should live in a pocket universe that starts out with a small Hubble rate, thus favoring emergent and cyclic models. Pocket universes which undergo cycles are further preferred, because they produce habitable conditions repeatedly inside each pocket.

Eternal inflation is a seemingly generic consequence of theories that give rise to accelerated expansion of the universe and possess multiple vacuum states. Making predictions in an eternally inflating universe is notoriously difficult because one must compare infinite quantities, and a wide variety of regulating procedures yield radically different results. This is the measure problem of eternal inflation. In this paper, we analyze models of eternal inflation which allow for the possibility of cyclic bubble universes: in each bubble, standard cosmological evolution is re-played over and over again. Eternal inflation can generically arise in cyclic models that include a dark energy dominated phase. In such models, several problematic consequences of standard regulating procedures, such as the youngness and Boltzmann Brain problems, are substantially alleviated. We discuss the implications for making predictions in cyclic models, as well as some general implications for understanding the measure problem in eternal inflation.

It has recently been argued by Copeland et al. that in eleven dimensions two orbifold planes can collide and bounce in a regular way, even when the bulk metric is perturbed away from Milne spacetime to a Kasner solution. In this paper, we point out that as a consequence the global "phoenix" structure of the cyclic universe is significantly enriched. Spatially separated regions, with different density fluctuation amplitudes as well as different non-gaussian characteristics, are all physically realized. Those regions containing by far the most structure are specified by a fluctuation amplitude of Q ~ 10^{-4.5} and local non-gaussianity parameters f_{NL} ~ O(+/- 10) and g_{NL} ~ O(-10^3), in agreement with current observations.

Galileon theories are of considerable interest since they allow for stable violations of the null energy condition. Since such violations could have occurred during a high-energy regime in the history of our universe, we are motivated to study supersymmetric extensions of these theories. This is carried out in this paper, where we construct generic classes of N=1 supersymmetric Galileon Lagrangians. They are shown to admit non-equivalent stress-energy tensors and, hence, vacua manifesting differing conditions for violating the null energy condition. The temporal and spatial fluctuations of all component fields of the supermultiplet are analyzed and shown to be stable on a large number of such backgrounds. In the process, we uncover a surprising connection between conformal Galileon and ghost condensate theories, allowing for a deeper understanding of both types of theories.

Cosmological models involving a bounce from a contracting to an expanding universe can address the standard cosmological puzzles and generate "primordial" density perturbations without the need for inflation. Some such models, in particular the ekpyrotic and cyclic models that we focus on, fit rather naturally into string theory. We discuss a number of topics related to these models: the reasoning that leads to the ekpyrotic phase, the predictions for upcoming observations, the differences between singular and non-singular models of the bounce as well as the predictive and explanatory power offered by these models.

We show how to construct supersymmetric actions for higher-derivative scalar field theories of the form P(X,phi), within the context of d=4, N=1 supersymmetry. This construction is of general use, and is applied to write supersymmetric versions of the Dirac-Born-Infeld action. Our principal application of this formalism is to construct the supersymmetric extension of the ghost condensate. This allows us to study the interplay between supersymmetry, time-dependent backgrounds and violations of the null energy condition.

In inflationary models, the predicted amplitude of primordial density perturbations Q is much larger than the observed value (~10^{-5}) for natural choices of parameters. To explain the requisite exponential fine-tuning, anthropic selection is often invoked, especially in cases where microphysics is expected to produce a complex energy landscape. By contrast, we find examples of ekpyrotic models based on heterotic M-theory for which dynamical selection naturally favors the observed value of Q.

Ekpyrotic models and their cyclic extensions solve the standard cosmological flatness, horizon and homogeneity puzzles by postulating a slowly contracting phase of the universe prior to the big bang. This ekpyrotic phase also manages to produce a nearly scale-invariant spectrum of scalar density fluctuations, but, crucially, with significant non-gaussian corrections. In fact, some versions of ekpyrosis are on the borderline of being ruled out by observations, while, interestingly, the best-motivated models predict levels of non-gaussianity that will be measurable by near-future experiments. Here, we review these predictions in detail, and comment on their implications.

Georges Lemaitre introduced the term "phoenix universe" to describe an oscillatory cosmology with alternating periods of gravitational collapse and expansion. This model is ruled out observationally because it requires a supercritical mass density and cannot accommodate dark energy. However, a new cyclic theory of the universe has been proposed that evades these problems. In a recent elaboration of this picture, almost the entire universe observed today is fated to become entrapped inside black holes, but a tiny region will emerge from these ashes like a phoenix to form an even larger smooth, flat universe filled with galaxies, stars, planets, and, presumably, life. Survival depends crucially on dark energy and suggests a reason why its density is small and positive today.

Using the covariant formalism, we derive the equations of motion for adiabatic and entropy perturbations at third order in perturbation theory for cosmological models involving two scalar fields. We use these equations to calculate the trispectrum of ekpyrotic and cyclic models in which the density perturbations are generated via the entropic mechanism. In these models, the conversion of entropy into curvature perturbations occurs just before the big bang, either during the ekpyrotic phase or during the subsequent kinetic energy dominated phase. In both cases, we find that the non-linearity parameters f_{NL} and g_{NL} combine to leave a very distinct observational imprint.

Non-gaussianity in the microwave background radiation is bound to play a key role in giving us clues about the physics of the very early universe. However, the associated calculations, at second and even third order in perturbation theory, tend to be complicated to the point of obscuring simple underlying physical processes. In this note, we present a simple analytic procedure for approximating the non-linearity parameters f_{NL} and g_{NL} for cyclic models in which the cosmological perturbations are generated via the entropic mechanism. Our approach is quick, physically transparent and agrees well with the results of numerical calculations.

It has been pointed out by several groups that ekpyrotic and cyclic models generate significant non-gaussianity. In this paper, we present a physically intuitive, semi-analytic estimate of the bispectrum. We show that, in all such models, there is an intrinsic contribution to the non-gaussianity parameter f_{NL} that is determined by the geometric mean of the equation of state w_{ek} during the ekpyrotic phase and w_{c} during the phase that curvature perturbations are generated and whose value is O(100) or more times the intrinsic value predicted by simple slow-roll inflationary models, f_{NL}^{intrinsic} = O(0.1). Other contributions to f_{NL}, which we also estimate, can increase |f_{NL}| but are unlikely to decrease it significantly, making non-gaussianity a useful test of these models. Furthermore, we discuss a predicted correlation between the non-gaussianity and scalar spectral index that sharpens the test.

We analyze the non-gaussian density perturbations generated in ekpyrotic/cyclic models based on heterotic M-theory. In this picture, two scalar fields produce nearly scale-invariant entropic perturbations during an ekpyrotic phase that are converted into curvature modes {\it after the ekpyrotic phase is complete} and just before the big bang. Both intrinsic non-linearity in the entropy perturbation and the conversion process contribute to non-gaussianity. The range of the non-gaussianity parameter $f_{NL}$ depends on how gradual the conversion process is and the steepness of the scalar field potential during the ekpyrotic phase. Although a wider range is possible, in principle, natural values of the ekpyrotic parameters combined with a gradual conversion process lead to values of $-60 \lesssim f_{NL} \lesssim +80$, typically much greater than slow-roll inflation but within the current observational bounds.

Ekpyrotic and cyclic cosmologies provide theories of the very early and of the very late universe. In these models, the big bang is described as a collision of branes - and thus the big bang is not the beginning of time. Before the big bang, there is an ekpyrotic phase with equation of state w=P/rho >> 1 (where P is the average pressure and rho the average energy density) during which the universe slowly contracts. This phase resolves the standard cosmological puzzles and generates a nearly scale-invariant spectrum of cosmological perturbations containing a significant non-gaussian component. At the same time it produces small-amplitude gravitational waves with a blue spectrum. The dark energy dominating the present-day cosmological evolution is reinterpreted as a small attractive force between our brane and a parallel one. This force eventually induces a new ekpyrotic phase and a new brane collision, leading to the idea of a cyclic universe. This review discusses the detailed properties of these models, their embedding in M-theory and their viability, with an emphasis on open issues and observational signatures.

A new and powerful probe of the origin and evolution of structures in the Universe has emerged and been actively developed over the last decade. In the coming decade, non-Gaussianity, i.e., the study of non-Gaussian contributions to the correlations of cosmological fluctuations, will become an important probe of both the early and the late Universe. Specifically, it will play a leading role in furthering our understanding of two fundamental aspects of cosmology and astrophysics: (i) the physics of the very early universe that created the primordial seeds for large-scale structures, and (ii) the subsequent growth of structures via gravitational instability and gas physics at later times. To date, observations of fluctuations in the Cosmic Microwave Background (CMB) and the Large-Scale Structure of the Universe (LSS) have focused largely on the Gaussian contribution as measured by the two-point correlations (or the power spectrum) of density fluctuations. However, an even greater amount of information is contained in non-Gaussianity and a large discovery space therefore still remains to be explored. Many observational probes can be used to measure non-Gaussianity, including CMB, LSS, gravitational lensing, Lyman-alpha forest, 21-cm fluctuations, and the abundance of rare objects such as clusters of galaxies and high-redshift galaxies. Not only does the study of non-Gaussianity maximize the science return from a plethora of present and future cosmological experiments and observations, but it also carries great potential for important discoveries in the coming decade.

In cyclic universe models based on a single scalar field (e.g., the radion determining the distance between branes in M-theory), virtually the entire universe makes it through the ekpyrotic smoothing and flattening phase, bounces, and enters a new epoch of expansion and cooling. This stable evolution cannot occur, however, if scale-invariant curvature perturbations are produced by the entropic mechanism because it requires two scalar fields (e.g., the radion and the Calabi-Yau dilaton) evolving along an unstable classical trajectory. In fact, we show here that an overwhelming fraction of the universe fails to make it through the ekpyrotic phase; nevertheless, a sufficient volume survives and cycling continues forever provided the dark energy phase of the cycle lasts long enough, of order a trillion years. Two consequences are a new role for dark energy and a global structure of the universe radically different from that of eternal inflation.

We discuss the moduli space approximation for heterotic M-theory, both for the minimal case of two boundary branes only, and when a bulk brane is included. The resulting effective actions may be used to describe the cosmological dynamics in the regime where the branes are moving slowly, away from singularities. We make use of the recently derived colliding branes solution to determine the global structure of moduli space, finding a boundary at which the trajectories undergo a hard wall reflection. This has important consequences for the allowed moduli space trajectories, and for the behaviour of cosmological perturbations in the model.

We analyze a general mechanism for producing a nearly scale-invariant spectrum of cosmological curvature perturbations during a contracting phase preceding a big bang, that can be entirely described using 4d effective field theory. The mechanism, based on first producing entropic perturbations and then converting them to curvature perturbations, can be naturally incorporated in cyclic and ekpyrotic models in which the big bang is modelled as a brane collision, as well as other types of cosmological models with a pre-big bang phase. We show that the correct perturbation amplitude can be obtained and that the scalar spectral tilt n tends to range from slightly blue to red, with 0.97 < n < 1.02 for the simplest models, a range compatible with current observations but shifted by a few per cent towards the blue compared to the prediction of the simplest, large-field inflationary models.

We study the collision of two flat, parallel end-of-the-world branes in heterotic M-theory. By insisting that there is no divergence in the Riemann curvature as the collision approaches, we are able to single out a unique solution possessing the local geometry of (2d compactified Milne)/Z_2 x R_3, times a finite-volume Calabi-Yau manifold in the vicinity of the collision. At a finite time before and after the collision, a second type of singularity appears momentarily on the negative-tension brane, representing its bouncing off a zero of the bulk warp factor. We find this singularity to be remarkably mild and easily regularised. The various different cosmological solutions to heterotic M-theory previously found by other authors are shown to merely represent different portions of a unique flat cosmological solution to heterotic M-theory.

We show how the bosonic sector of the radion supermultiplet plus d=4, N=1 supergravity emerge from a consistent braneworld Kaluza-Klein reduction of D=5 M--theory. The radion and its associated pseudoscalar form an SL(2,R)/U(1) nonlinear sigma model. This braneworld system admits its own brane solution in the form of a 2-supercharge supersymmetric string. Requiring this to be free of singularities leads to an SL(2,Z) identification of the sigma model target space. The resulting radion mode has a minimum length; we suggest that this could be used to avoid the occurrence of singularities in brane-brane collisions. We discuss possible supersymmetric potentials for the radion supermultiplet and their relation to cosmological models such as the cyclic universe or hybrid inflation.