
Gravitational lensing of the CMB is a valuable cosmological signal that
correlates to tracers of largescale structure and acts as a important source
of confusion for primordial $B$mode polarization. Stateoftheart lensing
reconstruction analyses use quadratic estimators, which are easily applicable
to data. However, these estimators are known to be suboptimal, in particular
for polarization, and large improvements are expected to be possible for high
signaltonoise polarization experiments. We develop a method and numerical
code, $\rm{LensIt}$, that is able to find efficiently the most probable lensing
map, introducing no significant approximations to the lensed CMB likelihood,
and applicable to beamed and masked data with inhomogeneous noise. It works by
iteratively reconstructing the primordial unlensed CMB using a deflection
estimate and its inverse, and removing residual lensing from these maps with
quadratic estimator techniques. Roughly linear computational cost is maintained
due to fast convergence of iterative searches, combined with the local nature
of lensing. The method achieves the maximal improvement in signal to noise
expected from analytical considerations on the unmasked parts of the sky.
Delensing with this optimal map leads to forecast tensortoscalar ratio
parameter errors improved by a factor $\simeq 2 $ compared to the quadratic
estimator in a CMB stage IV configuration.

The spectral index of scalar perturbations is an important observable that
allows us to learn about inflationary physics. In particular, a detection of a
significant deviation from a constant spectral index could enable us to rule
out the simplest class of inflation models. We investigate whether future
observations could rule out canonical singlefield slowroll inflation given
the parameters allowed by current observational constraints. We find that
future measurements of a constant running (or running of the running) of the
spectral index over currently available scales are unlikely to achieve this.
However, there remains a large region of parameter space (especially when
considering the running of the running) for falsifying the assumed class of
slowroll models if future observations accurately constrain a much wider range
of scales.

Lensing of the CMB is an important effect, and is usually modelled by
remapping the unlensed CMB fields by a lensing deflection. However the lensing
deflections also change the photon path so that the emission angle is no longer
orthogonal to the background lastscattering surface. We give the first
calculation of the emissionangle corrections to the standard lensing
approximation from dipole (Doppler) sources for temperature and quadrupole
sources for temperature and polarization. We show that while the corrections
are negligible for the temperature and Emode polarization, additional
largescale Bmodes are produced with a white spectrum that dominates those
from postBorn field rotation (curl lensing). On large scales about one percent
of the total lensinginduced Bmode amplitude is expected to be due to this
effect. However, the photon emission angle does remain orthogonal to the
perturbed lastscattering surface due to time delay, and half of the
largescale emissionangle B modes cancel with B modes from time delay to give
a total contribution of about half a percent. While not important for planned
observations, the signal could ultimately limit the ability of delensing to
reveal low amplitudes of primordial gravitational waves.We also derive the
rotation of polarization due to multiple deflections between emission and
observation. The rotation angle is of quadratic order in the deflection angle,
and hence negligibly small: polarization typically rotates by less than an
arcsecond, orders of magnitude less than a smallscale image rotates due to
postBorn field rotation (which is quadratic in the shear). The fieldrotation
B modes dominate the other effects on small scales.

I calculate a hybrid crosspower spectrum estimator from the WMAP 5year CMB
temperature maps, discuss the goodness of fit, and then constrain cosmological
parameters. The spectrum and results are generally consistent with previous
results, though the power spectrum error bars are slightly smaller and there
are small shifts at high ell. The small improvement in error bars is obtained
at very low numerical cost but does not significantly improve parameter
constraints. I discuss the accuracy of the likelihood model and how constraints
on the optical depth translate into constraints on the reionization history
allowing for helium reionization. In the appendices I propose a simple
reionization parameterization that determines the history in terms of a
midpoint reionization redshift, and suggest a new likelihood approximation for
chisquaredlike distributions with varying skewness.

Lensing of the CMB is now a welldeveloped probe of largescale clustering
over a broad range of redshifts. By exploiting the nonGaussian imprints of
lensing in the polarization of the CMB, the CORE mission can produce a clean
map of the lensing deflections over nearly the fullsky. The number of highS/N
modes in this map will exceed current CMB lensing maps by a factor of 40, and
the measurement will be samplevariance limited on all scales where linear
theory is valid. Here, we summarise this mission product and discuss the
science that it will enable. For example, the summed mass of neutrinos will be
determined to an accuracy of 17 meV combining CORE lensing and CMB twopoint
information with contemporaneous BAO measurements, three times smaller than the
minimum total mass allowed by neutrino oscillations. In the search for Bmode
polarization from primordial gravitational waves with CORE, lensinduced
Bmodes will dominate over instrument noise, limiting constraints on the
gravitational wave power spectrum amplitude. With lensing reconstructed by
CORE, one can "delens" the observed polarization internally, reducing the
lensing Bmode power by 60%. This improves to 70% by combining lensing and CIB
measurements from CORE, reducing the error on the gravitational wave amplitude
by 2.5 compared to no delensing (in the null hypothesis). Lensing measurements
from CORE will allow calibration of the halo masses of the 40000 galaxy
clusters that it will find, with constraints dominated by the clean
polarizationbased estimators. CORE can accurately remove Galactic emission
from CMB maps with its 19 frequency channels. We present initial findings that
show that residual Galactic foreground contamination will not be a significant
source of bias for lensing power spectrum measurements with CORE. [abridged]

Lensing of the CMB is affected by postBorn lensing, producing corrections to
the convergence power spectrum and introducing field rotation. We show
numerically that the lensing convergence power spectrum is affected at the
$\lesssim 0.2\%$ level on accessible scales, and that this correction and the
field rotation are negligible for observations with arcminute beam and noise
levels $\gtrsim 1 \mu {\text{K}}\,{\text{arcmin}} $. The field rotation
generates $\sim 2.5\%$ of the total lensing Bmode polarization amplitude
($0.2\%$ in power on small scales), but has a blue spectrum on large scales,
making it highly subdominant to the convergence B modes on scales where they
are a source of confusion for the signal from primordial gravitational waves.
Since the postBorn signal is nonlinear, it also generates a bispectrum with
the convergence. We show that the postBorn contributions to the bispectrum
substantially change the shape predicted from largescale structure
nonlinearities alone, and hence must be included to estimate the expected
total signal and impact of bispectrum biases on CMB lensing reconstruction
quadratic estimators and other observables. The fieldrotation power spectrum
only becomes potentially detectable for noise levels $\ll 1 \mu
{\text{K}}\,{\text{arcmin}}$, but its bispectrum with the convergence may be
observable at $\sim 3\sigma$ with Stage IV observations. Rotationinduced and
convergenceinduced B modes are slightly correlated by the bispectrum, and the
bispectrum also produces additional contributions to the lensed BB power
spectrum.

Observed CMB anisotropies are lensed, and the lensed power spectra can be
calculated accurately assuming the lensing deflections are Gaussian. However,
the lensing deflections are actually slightly nonGaussian due to both
nonlinear largescale structure growth and postBorn corrections. We calculate
the leading correction to the lensed CMB power spectra from the
nonGaussianity, which is determined by the lensing bispectrum. The
lowestorder result gives $\sim 0.3\%$ corrections to the BB and EE
polarization spectra on smallscales, however we show that the effect on EE is
reduced by about a factor of two by higherorder Gaussian lensing smoothing,
rendering the total effect safely negligible for the foreseeable future. We
give a simple analytic model for the signal expected from skewness of the
largescale lensing field; the effect is similar to a net demagnification and
hence a small change in acoustic scale (and therefore out of phase with the
dominant lensing smoothing that predominantly affects the peaks and troughs of
the power spectrum).

We present a first internal delensing of CMB maps, both in temperature and
polarization, using the public foregroundcleaned (SMICA) Planck 2015 maps.
After forming quadratic estimates of the lensing potential, we use the
corresponding displacement field to undo the lensing on the same data. We build
differences of the delensed spectra to the original data spectra specifically
to look for delensing signatures. After taking into account reconstruction
noise biases in the delensed spectra, we find an expected sharpening of the
power spectrum acoustic peaks with a delensing efficiency of $29\,\%$ ($TT$)
$25\,\%$ ($TE$) and $22\,\%$ ($EE$). The detection significance of the
delensing effects is very high in all spectra: $12\,\sigma$ in $EE$
polarization; $18\,\sigma$ in $TE$; and $20\,\sigma$ in $TT$. The null
hypothesis of no lensing in the maps is rejected at $26\,\sigma$. While direct
detection of the power in lensing $B$modes themselves is not possible at high
significance at Planck noise levels, we do detect (at $4.5\,\sigma$ under the
null hypothesis) delensing effects in the $B$mode map, with $7\,\%$ reduction
in lensing power. Our results provide a first demonstration of polarization
delensing, and generally of internal CMB delensing, and stand in agreement with
the baseline $\Lambda$CDM Planck 2015 cosmology expectations.

We forecast the scientific capabilities to improve our understanding of
cosmic inflation of CORE, a proposed CMB space satellite submitted in response
to the ESA fifth call for a mediumsize mission opportunity. The CORE satellite
will map the CMB anisotropies in temperature and polarization in 19 frequency
channels spanning the range 60600 GHz. CORE will have an aggregate noise
sensitivity of $1.7 \mu$K$\cdot \,$arcmin and an angular resolution of 5' at
200 GHz. We explore the impact of telescope size and noise sensitivity on the
inflation science return by making forecasts for several instrumental
configurations. This study assumes that the lower and higher frequency channels
suffice to remove foreground contaminations and complements other related
studies of component separation and systematic effects, which will be reported
in other papers of the series "Exploring Cosmic Origins with CORE." We forecast
the capability to determine key inflationary parameters, to lower the detection
limit for the tensortoscalar ratio down to the $10^{3}$ level, to chart the
landscape of single field slowroll inflationary models, to constrain the epoch
of reheating, thus connecting inflation to the standard radiationmatter
dominated Big Bang era, to reconstruct the primordial power spectrum, to
constrain the contribution from isocurvature perturbations to the $10^{3}$
level, to improve constraints on the cosmic string tension to a level below the
presumptive GUT scale, and to improve the current measurements of primordial
nonGaussianities down to the $f_{NL}^{\rm local} < 1$ level. For all the
models explored, CORE alone will improve significantly on the present
constraints on the physics of inflation. Its capabilities will be further
enhanced by combining with complementary future cosmological observations.

We forecast the main cosmological parameter constraints achievable with the
CORE space mission which is dedicated to mapping the polarisation of the Cosmic
Microwave Background (CMB). CORE was recently submitted in response to ESA's
fifth call for mediumsized mission proposals (M5). Here we report the results
from our presubmission study of the impact of various instrumental options, in
particular the telescope size and sensitivity level, and review the great,
transformative potential of the mission as proposed. Specifically, we assess
the impact on a broad range of fundamental parameters of our Universe as a
function of the expected CMB characteristics, with other papers in the series
focusing on controlling astrophysical and instrumental residual systematics. In
this paper, we assume that only a few central CORE frequency channels are
usable for our purpose, all others being devoted to the cleaning of
astrophysical contaminants. On the theoretical side, we assume LCDM as our
general framework and quantify the improvement provided by CORE over the
current constraints from the Planck 2015 release. We also study the joint
sensitivity of CORE and of future Baryon Acoustic Oscillation and Large Scale
Structure experiments like DESI and Euclid. Specific constraints on the physics
of inflation are presented in another paper of the series. In addition to the
six parameters of the base LCDM, which describe the matter content of a
spatially flat universe with adiabatic and scalar primordial fluctuations from
inflation, we derive the precision achievable on parameters like those
describing curvature, neutrino physics, extra light relics, primordial helium
abundance, dark matter annihilation, recombination physics, variation of
fundamental constants, dark energy, modified gravity, reionization and cosmic
birefringence. (ABRIDGED)

CMB and lensing reconstruction power spectra are powerful probes of
cosmology. However they are correlated, since the CMB power spectra are lensed
and the lensing reconstruction is constructed using CMB multipoles. We perform
a full analysis of the auto and crosscovariances, including polarization
power spectra and minimum variance lensing estimators, and compare with
simulations of idealized future CMBS4 observations. Covariances sourced by
fluctuations in the unlensed CMB and instrumental noise can largely be removed
by using a realizationdependent subtraction of lensing reconstruction noise,
leaving a relatively simple covariance model that is dominated by
lensinginduced terms and well described by a small number of principal
components. The correlations between the CMB and lensing power spectra will be
detectable at the level of $\sim 5\sigma$ for a CMBS4 mission, and neglecting
those could underestimate some parameter error bars by several tens of percent.
However we found that the inclusion of external priors or data sets to estimate
parameter error bars can make the impact of the correlations almost negligible.

Negative absolute temperatures (NAT) are an exotic thermodynamical
consequence of quantum physics which has been known since the 1950's (having
been achieved in the lab on a number of occasions). Recently, the work of Braun
et al (2013) has rekindled interest in negative temperatures and hinted at a
possibility of using NAT systems in the lab as dark energy analogues. This
paper goes one step further, looking into the cosmological consequences of the
existence of a NAT component in the Universe. NATdominated expanding Universes
experience a borderline phantom expansion ($w<1$) with no Big Rip, and their
contracting counterparts are forced to bounce after the energy density becomes
sufficiently large. Both scenarios might be used to solve horizon and flatness
problems analogously to standard inflation and bouncing cosmologies. We discuss
the difficulties in obtaining and ending a NATdominated epoch, and possible
ways of obtaining density perturbations with an acceptable spectrum.

Percentlevel measurements of the comoving baryon acoustic scale standard
ruler can be used to break degeneracies in parameter constraints from the CMB
alone. The sound horizon at the epoch of baryon drag is often used as a proxy
for the scale of the peak in the matter density correlation function, and can
conveniently be calculated quickly for different cosmological models. However,
the measurements are not directly constraining this scale, but rather a
measurement of the full correlation function, which depends on the detailed
evolution through decoupling. We assess the level of reliability of parameter
constraints based on a simple approximation of the acoustic scale compared to a
more direct determination from the full numerical twopoint correlation
function. Using a fiveparameter fitting technique similar to recent BAO data
analyses, we find that for standard {\Lambda}CDM models and extensions with
massive neutrinos and additional relativistic degrees of freedom, the
approximation is at better than 0.15% for most parameter combinations varying
over reasonable ranges.

Detailed measurements of the CMB lensing signal are an important scientific
goal of ongoing groundbased CMB polarization experiments, which are mapping
the CMB at high resolution over small patches of the sky. In this work we
simulate CMB polarization lensing reconstruction for the $EE$ and $EB$
quadratic estimators with currentgeneration noise levels and resolution, and
show that without boundary effects the known and expected zeroth and first
order $N^{(0)}$ and $N^{(1)}$ biases provide an adequate model for nonsignal
contributions to the lensing power spectrum estimators. Small sky areas present
a number of additional challenges for polarization lensing reconstruction,
including leakage of $E$ modes into $B$ modes. We show how simple windowed
estimators using filtered pure$B$ modes can greatly reduce the maskinduced
meanfield lensing signal and reduce variance in the estimators. This provides
a simple method (used with recent observations) that gives an alternative to
more optimal but expensive inversevariance filtering.

We compute the bispectrum induced in the cosmic microwave background (CMB)
temperature and polarisation by the evolution of the primordial density
perturbations using the secondorder Boltzmann code SONG. We show that adding
polarisation increases the signaltonoise ratio by a factor four with respect
to temperature alone and we estimate the observability of this intrinsic
bispectrum and the bias it induces on measurements of primordial
nonGaussianity. When including all physical effects except the latetime
nonlinear evolution, we find for the intrinsic bispectrum a signaltonoise of
$S/N=3.8,\,2.9,\,1.6$ and $0.5$ for, respectively, an ideal experiment with an
angular resolution of $\ell_\text{max}=3000$, the proposed CMB surveys PRISM
and COrE, and Planck's polarised data; the bulk of this signal comes from the
$E$polarisation and from squeezed configurations. We discuss how CMB lensing
is expected to reduce these estimates as it suppresses the bispectrum for
squeezed configurations and contributes to the noise in the estimator. We find
that the presence of the intrinsic bispectrum will bias a measurement of
primordial nonGaussianity of local type by $f_{NL}^\text{bias}=0.66$ for an
ideal experiment with $\ell_\text{max}=3000$. Finally, we verify the robustness
of our results by reproducing the analytical approximation for the
squeezedlimit bispectrum in the general polarised case.

Rayleigh scattering from neutral hydrogen during and shortly after
recombination causes the CMB anisotropies to be significantly frequency
dependent at high frequencies. This may be detectable with Planck, and would be
a strong signal in any future spacebased CMB missions. The later peak of the
Rayleigh visibility compared to Thomson scattering gives an increased
largescale CMB polarization signal that is a greater than 4% effect for
observed frequencies greater than 500GHz. There is a similar magnitude
suppression on small scales from additional damping. Due to strong correlation
between the Rayleigh and primary signal, measurement of the Rayleigh component
is limited by noise and foregrounds, not cosmic variance of the primary CMB,
and should observable over a wide range of angular scales at frequencies
between roughly 200GHz and 800GHz. I give new numerical calculations of the
temperature and polarization power spectra, and show that future CMB missions
could measure the temperature Rayleigh crossspectrum at high precision, detect
the polarization from Rayleigh scattering, and also accurately determine the
crossspectra between the Rayleigh temperature signal and primary polarization.
The Rayleigh scattering signal may provide a powerful consistency check on
recombination physics. In principle it can be used to measure additional
horizonscale primordial perturbation modes at recombination, and distinguish a
significant tensor mode Bpolarization signal from gravitational lensing at the
power spectrum level.

Gravitational lensing provides a significant source of cosmological
information in modern CMB parameter analyses. It is measured in both the power
spectrum and trispectrum of the temperature fluctuations. These observables are
often treated as independent, although as they are both determined from the
same map this is impossible. In this paper, we perform a rigorous analysis of
the covariance between lensing power spectrum and trispectrum analyses. We find
two dominant contributions coming from: (i) correlations between the
disconnected noise bias in the trispectrum measurement and sample variance in
the temperature power spectrum; and (ii) sample variance of the lenses
themselves. The former is naturally removed when the dominant N0 Gaussian bias
in the reconstructed deflection spectrum is dealt with via a partially
datadependent correction, as advocated elsewhere for other reasons. The
remaining lenscosmicvariance contribution is easily modeled but can safely be
ignored for a Plancklike experiment, justifying treating the two observable
spectra as independent. We also test simple likelihood approximations for the
deflection power spectrum, finding that a Gaussian with a parameterindependent
covariance performs well.

Physical parameters are often constrained from the data likelihoods using
sampling methods. Changing some parameters can be much more computationally
expensive (`slow') than changing other parameters (`fast parameters'). I
describe a method for decorrelating fast and slow parameters so that parameter
sampling in the full space becomes almost as efficient as sampling in the slow
subspace when the covariance is well known and the distributions are simple.
This gives a large reduction in computational cost when there are many fast
parameters. The method can also be combined with a fast 'dragging' method
proposed by Neal (2005) that can be more robust and efficient when parameters
cannot be fully decorrelated a priori or have more complicated dependencies. I
illustrate these methods for the case of cosmological parameter estimation
using data likelihoods from the Planck satellite observations with dozens of
fast nuisance parameters, and demonstrate a speed up by a factor of five or
more. In more complicated cases, especially where the fast subspace is very
fast but complex or highly correlated, the fastslow sampling methods can in
principle give arbitrarily large performance gains. The new samplers are
implemented in the latest version of the publicly available CosmoMC code.

We study how to recover the full 3D clustering information of P(\vec{k},z),
including redshift space distortions (RSD), from 2D tomography using the
angular auto and cross spectra of different redshift bins C_\ell(z,z'). We
focus on quasilinear scales where the minimum scale \lambda_{min} or
corresponding maximum wavenumber k_{max}= 2\pi/\lambda_{min} is targeted to be
between k_{max}={0.050.2} h/Mpc. For spectroscopic surveys, we find that we
can recover the full 3D clustering information when the redshift bin width
\Delta z used in the 2D tomography is similar to the targeted minimum scale,
i.e. \Delta z ~ {0.60.8} \lambda_{min} H(z)/c which corresponds to \Delta z ~
0.010.05 for z<1. This value of \Delta z is optimal in the sense that larger
values of \Delta z lose information, while smaller values violate our minimum
scale requirement. For a narrowband photometric survey, with photoz error
\sigma_z=0.004, we find almost identical results to the spectroscopic survey
because the photoz error is smaller than the optimal bin width \sigma_z<\Delta
z. For a typical broadband photometric survey with \sigma_z=0.1, we have that
\sigma_z>\Delta z and most radial information is intrinsically lost. The
remaining information can be recovered from the 2D tomography if we use \Delta
z ~ 2\sigma_z. While 3D and 2D analysis are shown here to be equivalent, the
advantage of using angular positions and redshifts is that we do not need a
fiducial cosmology to convert to 3D coordinates. This avoids assumptions and
marginalization over the fiducial model. In addition, it becomes straight
forward to combine RSD, clustering and weak lensing in 2D space.

Primordial magnetic fields could provide an explanation for the galactic
magnetic fields observed today, in which case they may also leave interesting
signals in the CMB and the smallscale matter power spectrum. We discuss how to
approximately calculate the important nonlinear magnetic effects within the
guise of linear perturbation theory, and calculate the matter and CMB power
spectra including the SZ contribution. We then use various cosmological
datasets to constrain the form of the magnetic field power spectrum. Using
solely largescale CMB data (WMAP7, QUaD and ACBAR) we find a 95% CL on the
variance of the magnetic field at 1 Mpc of B_\lambda < 6.4 nG. When we include
SPT data to constrain the SZ effect, we find a revised limit of B_\lambda < 4.1
nG. The addition of SDSS Lymanalpha data lowers this limit even further,
roughly constraining the magnetic field to B_\lambda < 1.3 nG.

The smallscale CMB temperature we observe on the sky is modulated by
perturbations that were superhorizon at recombination, giving differential
focussing and lensing that generate a nonzero bispectrum even for singlefield
inflation where local physics is identical. Understanding this signal is
important for primordial nonGaussianity studies and also parameter constraints
from the CMB lensing bispectrum signal. Because of cancellations individual
effects can appear larger or smaller than they are in total, so a full analysis
may be required to avoid biases. I relate angular scales on the sky to physical
scales at recombination using the optical equations, and give fullsky results
for the largescale adiabatic temperature bispectrum from Ricci focussing
(expansion of the ray bundle), Weyl lensing (convergence and shear), and
temperature redshift modulations of smallscale power. The delta N expansion of
the beam is described by the constant temperature 3curvature, and gives a
nearlyobservable version of the consistency relation prediction from
singlefield inflation. I give approximate arguments to quantify the likely
importance of dynamical effects, and argue that they can be neglected for
modulation scales l <~ 100, which is sufficient for lensing studies and also
allows robust tests of local primordial nonGaussianity using only the
largescale modulation modes. For accurate numerical results early and
latetime ISW effects must be accounted for, though I confirm that the
latetime nonlinear ReesSciama contribution is negligible compared to other
more important complications. The total corresponds to f_NL ~ 7 for Plancklike
temperature constraints and f_NL ~ 11 for cosmicvariance limited data to
lmax=2000. Temperature lensing bispectrum estimates are affected at the 0.2
sigma level by Ricci focussing, and up to 0.5 sigma with polarization.

Microwave background temperature and polarization observations are a powerful
way to constrain cosmological parameters if the likelihood function can be
calculated accurately. The temperature and polarization fields are correlated,
partial sky coverage correlates power spectrum estimators at different ell, and
the likelihood function for a theory spectrum given a set of observed
estimators is nonGaussian. An accurate analysis must model all these
properties. Most existing likelihood approximations are good enough for a
temperatureonly analysis, however they cannot reliably handle a
temperaturepolarization correlations. We give a new general approximation
applicable for correlated Gaussian fields observed on part of the sky. The
approximation models the nonGaussian form exactly in the ideal fullsky limit
and is fast to evaluate using a precomputed covariance matrix and set of power
spectrum estimators. We show with simulations that it is good enough to obtain
correct results at ell >~ 30 where an exact calculation becomes impossible. We
also show that some Gaussian approximations give reliable parameter constraints
even though they do not capture the shape of the likelihood function at each
ell accurately. Finally we test the approximations on simulations with
realistically anisotropic noise and asymmetric foreground mask.

Cosmological parameter constraints from the CMB power spectra alone suffer
several wellknown degeneracies. These degeneracies can be broken by numerical
artefacts and also a variety of physical effects that become quantitatively
important with highaccuracy data e.g. from the Planck satellite. We study
degeneracies in models with flat and nonflat spatial sections, nontrivial
dark energy and massive neutrinos, and investigate the importance of various
physical degeneracybreaking effects. We test the CAMB power spectrum code for
numerical accuracy, and demonstrate that the numerical calculations are
accurate enough for degeneracies to be broken mainly by true physical effects
(the integrated SachsWolfe effect, CMB lensing and geometrical and other
effects through recombination) rather than numerical artefacts. We quantify the
impact of CMB lensing on the power spectra, which inevitably provides
degeneracybreaking information even without using information in the
nonGaussianity. Finally we check the numerical accuracy of samplebased
parameter constraints using CAMB and CosmoMC. In an appendix we document recent
changes to CAMB's numerical treatment of massive neutrino perturbations, which
are tested along with other recent improvements by our degeneracy exploration
results.

Squeezed primordial nonGaussianity can strongly constrain earlyuniverse
physics, but it can only be observed on the CMB after it has been
gravitationally lensed. We give a new simple nonperturbative prescription for
accurately calculating the effect of lensing on any squeezed primordial
bispectrum shape, and test it with simulations. We give the generalization to
polarization bispectra, and discuss the effect of lensing on the trispectrum.
We explain why neglecting the lensing smoothing effect does not significantly
bias estimators of local primordial nonGaussianity, even though the change in
shape can be >~10%. We also show how tau_NL trispectrum estimators can be well
approximated by much simpler CMB temperature modulation estimators, and hence
that there is potentially a ~1030% bias due to very largescale lensing modes,
depending on the range of modulation scales included. Including dipole sky
modulations can halve the tau_NL error bar if kinematic effects can be
subtracted using known properties of the CMB temperature dipole. Lensing
effects on the g_NL trispectrum are small compared to the error bar. In
appendices we give the general result for lensing of any primordial bispectrum,
and show how any fullsky squeezed bispectrum can be decomposed into orthogonal
modes of distinct angular dependence.

I review what bispectra and trispectra look like in real space, in terms of
the sign of particular shaped triangles and tetrahedrons. Having an equilateral
density bispectrum of positive sign corresponds to having concentrated
overdensities surrounded by larger weaker underdensities. In 3D these are
concentrated density filaments, as expected in largescale structure. As the
shape changes from equilateral to flattened the concentrated overdensities
flatten into lines (3D planes). I then focus on squeezed bispectra, which can
be thought of as correlations of changes in smallscale power with largescale
fields, and discuss the general nonperturbative form of the squeezed
bispectrum and its angular dependence. A general trispectrum has tetrahedral
form and I show examples of what this can look like in real space. Squeezed
trispectra are of particular interest and come in two forms, corresponding to
largescale variance of smallscale power, and correlated modulations of an
equilateralform bispectrum. Flattened trispectra can be produced by linelike
features in 2D, for example from cosmic strings, and randomly located features
also give a nonGaussian signal. There are relationships between the squeezed
types of nonGaussianity, and also a useful interpretation in terms of
statistical anisotropy. I discuss the various possible physical origins of
cosmological nonGaussianities, both in terms of primordial perturbations and
latetime dynamical and geometric effects.