INFERENCE / Vol. 6, No.

Heart of Darkness
Subir Sarkar

Cosmologists are often in error, but never in doubt. The standard model of comology embodies a number of
—Lev Landau1 assumptions.

I
First, space-time is described by the Friedmann–
n the standard model of cosmology, about seventy Lemaître–Robertson–Walker (FLRW) metric:7
percent of the energy density of the universe—the dark
energy driving its accelerating rate of expansion—is 1. ds2 ≡ gμvxμxv = a2(η)[dη2 – dx2], a2(η)dη2 ≡ dt2,
described by Albert Einstein’s cosmological constant.2 In
this essay, I argue that the standard model of cosmology where gμv is the metric tensor, a is a scale factor de-
is wrong. This should come as no surprise. “The history scribing the expansion or contraction of space, and η is
of science,” Georges Lemaître remarked, “provides many conformal time. The conformal transformation denoted
instances of discoveries which have been made for reasons by a2 stretches space without distorting shapes, and the
which are no longer considered satisfactory.” It may be, he redshift z measures the wavelength of light from distant
added suggestively, “that the discovery of the cosmological objects. It reflects the scale factor at the time that the light
constant is such a case.”3 was emitted as a ratio to its current value: 1 + z ≡ a/a0.
Einstein published the general theory of relativity in Second, gravity is described by general relativity:
1915; and in 1917, he attempted to apply his theory to the
cosmos as a whole. The result was a universe that was 2. Rμv – Rgμv + 𝜆gμv = 8πGNTμv ,
either expanding or contracting, and, in any case, not
static. Dissatisfied with this implication, Einstein added a where the cosmological constant, 𝜆, acts as a repulsive
constant to his field equation, its repulsive force balanc- force that increases with distance, thus counteracting the
ing the otherwise attractive force of gravity.4 The result attractive force of gravitation.8
was a static, spherical, spatially closed but unstable uni- Third, the only contents of the universe are ideal fluids
verse. At almost the same time, Willem de Sitter provided possessing pressure p and energy density 𝜌, but neither
a maximally symmetric vacuum solution of Einstein’s field viscosity nor vorticity nor any other dissipative properties.
equations. This too was initially believed to be static. The With the publication of Einstein’s field equations, cos-
ensuing universe contained neither radiation nor matter, mologists originally argued that matter could be modeled
but, given its positive cosmological constant, it was obliged as a pressureless gas—dust, in effect. That is still their
to expand at an accelerating rate. It is de Sitter’s universe assumption with respect to dark matter. It is no longer a
that is effectively the basis of the current standard model, general assumption because the early universe was dom-
which was subsequently constructed by Alexander Fried- inated by relativistic radiation. Even earlier, the universe
mann and Lemaître.5 was hot enough to melt all particles of matter. A descrip-

A
tion of Tμv in terms of quantum field theory is obligatory.
standard model for cosmology is nothing new. The complicated relationships between the geometry
In Europe, Ptolemy’s geocentric theory held sway of space-time and the matter that it contains now admits
for nearly two thousand years. If its underlying of simplification:
assumptions had no physical basis, the theory was better
( ) = 8 G3 𝜌
2
ȧ π k Λ
than the competition.6 “Absolutely all phenomena, are in 3. H2 ≡ N m
– +
a a2 3
contradiction,” Ptolemy wrote, “to any of the alternate
notions that have been propounded.” Like the geocentric = H02[Ωm(1 + z)3 + Ωk(1 + z)2 + ΩΛ] ,
model, the underlying assumptions of the standard model
have no physical basis. Cosmologists still lack an under- where
standing of dark energy, and they do not know what is
driving the accelerating expansion of the universe. • Ωm ≡ 𝜌m /(3H02 / 8πGN),

1 / 10
 CRITICAL ESSAYS

• Ωk ≡ –k/(3H02 a02), and the universe is spatially flat. Attempts to weigh massive
• ΩΛ ≡ Λ /(3H02). clusters of galaxies, on the other hand, indicate that
Ωm = 0.3 ± 0.1—not enough to provide the critical density
This is the Friedmann–Lemaître equation. It describes for a flat universe. Adjusting yin to yang requires that
the rate of change of the Hubble parameter, H, as a function ΩΛ = 1 – Ωm – Ωk ≃ 0.7. The universe is apparently domi-
of the energy density of matter 𝜌m, the curvature of spatial nated by a cosmological constant with the value Λ ~ 2H02.
sections k, and the cosmological constant itself. A similar The scale of Λ is thus set by H0, which is measured to be
term also arises from the right-hand side of this equation ~70 km/sec/Mpc (i.e., h ≡ H0 /100 km/s/Mpc ≃ 0.7). Its
when Tμv describes quantum fields, a point first realized inverse, the Hubble radius H0−1 ≃ 3000/h Mpc, corresponds
by Wolfgang Pauli and Yakov Zeldovich. If all inertial to the tiny energy scale of ~10–42 GeV in fundamental phys-
observers must see the same vacuum state, ground-state ics units. Although neither fundamental nor a constant, it
quantum fluctuations behave like a cosmological constant enters into every cosmological measurement. Any infer-
with negative pressure: ence of Λ from the data in the FLRW framework will thus
yield a value of order H02.
4. Tμv = –⟨𝜌⟩fields gμv . Alarm bells ought to be ringing.
On the other hand, the acceleration of the expansion rate
It is this that enters in the Friedmann–Lemaître equa- is driven by the negative pressure of the quantum vacuum:
tion (3), which has been rewritten above in terms of the –pΛ = ρΛ ~ Λ/8πGN ~ (10–12 GeV)4, since 1/8πGN ≡ M2Planck
fractional contributions made to the total energy density ≃ (2.4 × 1018 GeV)2. This corresponds to a dark-energy scale
by matter, curvature, and the cosmological constant.9 of ~10–12 GeV. With respect to Λ, which is it to be: ~10–42
The effective cosmological constant, Λ, is the sum of GeV or ~10–12 GeV? Cosmological and quantum-field the-
these unrelated terms: oretic interpretations of Λ reveal a discrepancy of between
15 and 30 orders of magnitude, corresponding to a factor of
5. Λ = λ + 8πGN⟨𝜌⟩fields . 1060 to 10120 in the energy density.
Not surprisingly, this has been called “the worst theo-
The cosmic sum rule follows upon division by H02: retical prediction in the history of physics.”13
If dramatic, the statement is misleading. The vacuum
6. Ωm + Ωk + ΩΛ = 1. energy in quantum field theory cannot be formally cal-
culated because it is a super-renormalizable term in the
This simple equation encapsulates the standard FLRW Lagrangian. Only the difference in energy density between
cosmological model, and it is this equation that is used to two vacuum states can be computed. If there are changes
deduce the values of various cosmological parameters. in the vacuum state, further ambiguities arise in calculat-
Since Ωm is ostensibly comprised of cold—i.e., nonrelativis- ing the expected vacuum energy density. For astronomers,
tic—dark matter (CDM), it is called the ΛCDM model. Λ is just another cosmological parameter, similar to 𝜌m, the

I
matter density, or k, the curvature of spatial sections. From
n the late 1990s, observations of Type Ia (SNe Ia) the viewpoint of a quantum field theorist, Λ represents a
supernovae indicated that the expansion of the uni- very delicate balancing act between the bare λ derived
verse was speeding up. These stellar thermonuclear from general coordinate invariance and the divergent con-
explosions are expected to release a standard amount of tributions from various quantum fields. These must yield
visible energy, and so make for a standard candle in astron- an overall value of Λ that is consistent with cosmological
omy.10 Such supernovae are detectable up to cosmological observations.
distances and, in conjunction with the redshift, provide The less appreciated point is that in quantum field
a means for tracking the Hubble expansion rate beyond theory, all contributions to the vacuum energy can be
z ~ 1.11 The observed magnitude μ is a measure of the formally canceled order by order by adding appropriate
luminosity distance μ ≡ 5log10(dL / 10 pc), which is, in turn, counterterms to the Lagrangian. This would have no effect
related to various cosmological parameters by on any quantity measured in the laboratory. It is only when
the standard model of particle physics is supplemented
by gravity that the quantum vacuum can have a possible
7. dL = c
(1 + z)
√ΩkH0 (
sinn √Ωk ∫0
z

)
dz′H0 /H(z′) , effect on the universe as a whole.
When these implications were first realized in the
1930s, they were quickly swept under the rug. “[A]s is
where sinn → sinh for Ωk > 0 and sinn → sin for Ωk < 0 obvious from experience,” Pauli confidently remarked,
(1 parsec ≃ 3.3 lightyears). “the [zero-point energy] does not produce any gravita-
These observations indicate that 0.8Ωm – 0.6ΩΛ = –0.2 ± tional field.”14 This is of course true. If the zero-point
0.1.12 The typical angular scale of temperature fluctuations energy of the standard model were to gravitate, it would
in the cosmic microwave background shows that Ωk ≃ 0: have dominated the universe when it cooled to tempera-

2 / 10
 INFERENCE / Vol. 6, No. 4

tures of ~1015 K (⇒ ~100 GeV) at ~10–12s after the Big Bang. radiation bath of temperature T0 will measure an effective
Depending on the sign, this would have either sent the temperature that varies with respect to the direction of
universe off into an endless, exponentially fast expansion, motion by the angle θ :19
or caused it to immediately re-collapse. Pauli offered no
reason why the vacuum energy density should not gravi- 8. T(θ) = T0 √1 − β2 / (1 − βcosθ).
tate, and overlooked, or ignored, the obvious conflict with
general relativity. In Einstein’s theory, all forms of energy Since our peculiar velocity was estimated to be a few
must gravitate. hundred km/s, the amplitude of the dipole in the cosmic
This is the bone in our throat.15 microwave background temperature should then be 𝒟 = β

T
≃ 10–3. This predicted anisotropy was indeed detected soon
he oxford mathematician Edward Arthur Milne afterward.20 It has been measured most precisely using the
formulated the modern version of the cosmological Planck satellite.21 Its amplitude is (1.2336 ± 0.0004) × 10–3,
principle in 1933. “The Universe,” he wrote, “must implying that the sun is moving with respect to the cosmic
appear the same to all observers.”16 This assumption was, rest frame at 369.82 ± 0.11 km/s toward galactic longitude
in fact, implicit in models constructed a century ago. The and latitude: l(deg) = 264.021 ± 0.011, b(deg) = 48.253 ±
observed universe is not quite isotropic, and neither is it 0.005, which is in the constellation of Crater.
homogeneous. The interpretation of data in terms of the For this reason, the kinematic interpretation of the
FLRW model has proceeded under the weaker assumptions cosmic microwave background dipole has been widely
of statistical isotropy and homogeneity. Although there accepted. Cosmological data are now routinely corrected
are known inhomogeneities in the distribution of matter, by a special-relativity boost transforming the measured
homogeneity is restored when averaged on sufficiently large redshift and magnitudes of distant objects to the presump-
scales. Number counts of galaxies in spheres of increasing tively isotropic cosmic rest frame. In the cosmic microwave
radius r are said to have demonstrated a close approach to background frame, the large-scale averaged distribution of
homogeneity on spatial scales above ~70/h Mpc.17 matter is also assumed to be isotropic. The Friedmann–
For all that, we are manifestly not in the cosmic rest Lemaître equations can then be applied to the transformed
frame in which the universe is homogeneous and isotropic, magnitudes and redshifts.

T
and data can be analyzed according to the Friedmann–
Lemaître equation. If this were so, then all galaxies would hese assumptions are no longer tenable. Sev-
be receding from the earth at a rate proportional to their eral independent data sets now argue against the
distance, following Hubble’s law. But it has been known existence of a cosmic rest frame. At low redshift
since the 1960s that we have a peculiar, non-Hubble veloc- (z ≲ 0.1), all matter in our local supercluster of galaxies
ity due to local inhomogeneities. The idealized Hubble has a coherent bulk flow approximately aligned with the
flow can emerge only after sufficiently large scales are direction of the cosmic microwave background dipole: no
averaged and peculiar velocities have died away. This convergence to the cosmic rest frame is seen on scales as
scale should be of the same order as the scale on which the large as ~300/h Mpc. At high redshift (z > 1), the observed
distribution of matter, as traced by the visible galaxies, is dipole in the sky-distribution of distant radio sources and
sensibly homogeneous, around ~70/h Mpc. quasars is significantly larger than expected according
Since we are not in the cosmic rest frame, the cosmic to the kinematic interpretation of the cosmic microwave
microwave background cannot look isotropic. It ought background dipole. Phenomena are in conflict with the
to exhibit a pronounced dipole anisotropy. The expected cosmological principle. They directly challenge the claim
anisotropy is due to aberration. An isotropically distributed that the universe is dominated by vacuum energy, which
set of objects should acquire a dipole anisotropy—more rests on its assumed large-scale homogeneity and isotropy.
than average density in the direction of motion, less in the These are potentially paradigm-changing develop-
opposite direction. There is a second effect that boosts ments.
distortion: photons coming toward us from the direction Since the pioneering studies by the astronomer Vera
of our motion are Doppler-shifted to the blue, while the Rubin and her collaborators in the 1970s, it has been
photons coming from behind are Doppler-shifted to the known that our Local Group of galaxies, which includes
red. The strength of both effects should be proportional to Andromeda and the Magellanic Clouds, participates in a
β, the ratio of our velocity to the speed of light. This was fast coherent bulk flow.22 This departure from the uniform
first recognized by Dennis Sciama and John Stewart, who Hubble expansion happens because the local distribution
emphasized that, “If the microwave blackbody radiation is of matter is inhomogeneous. The flow would be expected to
both cosmological and isotropic, it will only be isotropic become negligible by ~100 Mpc, after which the averaged
to an observer who is at rest in the rest frame of distant universe becomes homogeneous. The measurement of
matter which last scattered the radiation.”18 Jim Peebles the cosmic microwave background dipole yields the sun’s
and David Wilkinson then calculated that an inertial peculiar velocity relative to the cosmic rest frame. The sun
observer moving with velocity β in an isotropic blackbody orbits around the Milky Way in nearly the opposite direc-

3 / 10
 CRITICAL ESSAYS

tion at ~200 km/s, and the Milky Way is moving toward sources normally too faint to be detected in a flux-limited
the center of the Local Group at ~40 km/s. All this sums survey should be boosted above the threshold if they are
to a net motion of the Local Group at 620 ± 15 km/s toward in the forward direction, and boosted below the threshold
l(deg) = 271.9 ± 2.0, b(deg) = 29.6 ± 1.4, which is not far from if they are behind. This enhances the distortion consider-
the direction of the cosmic microwave background dipole. ably.
This was ascribed to the attraction of a gigantic mass con- The predicted dipole anisotropy has amplitude
centration dubbed the Great Attractor, weighing ~5 × 1016
solar masses at a distance of ~65/h Mpc.23 9. 𝒟 = [2 + x(1 + α)]β ,
Subsequently, an infrared survey by the Infrared
Astronomical Satellite reached a depth of ~100/h Mpc. corresponding to a maximum change of about 0.5% in the
It showed that while the summed effects of the known density of radio sources that have on average ⟨α⟩ ~ 0.75
superclusters of galaxies could account for much of the and ⟨x⟩ ~ 1.29
observed bulk flow, there was still no convergence to The data available at that time were inadequate to per-
the cosmic rest frame. Infrared and X-ray surveys have form this test—as they noted. Measuring the expected
brought further evidence that the flow continues out to the dipole requires counting the density on the sky of at least
Shapley Supercluster at ~180/h Mpc.24 We confirmed this several hundred thousand sources at high redshift, in
using the Union 2 catalog of Type Ia supernovae to obtain order to adequately suppress random fluctuations.
distances to their host galaxies in order to perform tomog- The first catalog of radio sources became available only
raphy of the local velocity field. Using the same technique, in this millennium. It is the National Radio Astronomy
the Nearby Supernova Factory collaboration has shown Observatory (NRAO) Very Large Array (VLA) Sky Survey,
that the bulk flow continues even beyond Shapley, out to which mapped the entire sky north of –40 degrees decli-
~300 Mpc, thus requiring an even bigger inhomogeneity nation at 1.4 GHz. Surprisingly, the radio dipole was found
to drive it.25 A detailed map of local structures by Brent to be over twice the predicted value, although consistent
Tully and collaborators uses direct distance measure- with the cosmic microwave background dipole in direc-
ments to determine these peculiar velocities. It shows that tion.30 This exercise has been redone by many researchers
this motion is in fact coherent across the Laniakea Super- who have found similar results, albeit all of less than
cluster, in which we live.26 three sigma in significance.31 The anomaly has not been
So far as the universe has been mapped in detail, there accepted as genuine. Some researchers argue that it is due
is no convergence to the cosmic microwave background to unidentified systematics in the mapmaking or galactic
frame. contaminants such as nearby radio sources in the Milky

I
Way. Such concerns can be addressed by making suit-
n a review of these puzzling observations, the astron- able cuts on the minimum flux and by masking out the
omer James Gunn expressed a radical thought: “Most Milky Way. The latter also helps to eliminate a clustering
of the problem, it seems to me, would disappear if the dipole—the accidental proximity of a random fluctuation
[cosmic microwave background] did not, in fact, provide a in the source density. This was in fact guarded against in
rest frame.”27 some of the cited analyses by cross-correlating with other
A direct test of this had, in fact, been formulated in 1984 catalogs of nearby galaxies and excluding any sources in
when George Ellis and John Baldwin argued that if common.
It remains true that the distribution in redshift of the
the standard interpretation of the dipole anisotropy in radio sources is not directly measured but only inferred
the microwave background radiation as being due to our from their flux distribution. One cannot be certain that
peculiar velocity in a homogeneous isotropic universe is some of them do not happen to be accidentally nearby. One
correct, then radio-source number counts must show a might even wonder if it is just such a local cluster that is
similar anisotropy. Conversely, determination of a dipole also responsible for pulling us in the faster-than-expected
anisotropy in those counts determines our velocity rela- bulk flow that stretches out farther than expected.

T
tive to their rest frame; this velocity must agree with that
determined from the microwave background radiation o eliminate this possibility, it is necessary to
anisotropy.28 establish that sky-sources are cosmologically
distant. Doing so requires measuring their red-
Unlike the cosmic microwave background, radio shifts spectroscopically, rather than by more uncertain
sources have a nonthermal spectrum. Their flux density photometric estimates. This is a laborious process. To
Sv—the power emitted in units of W/m2/Hz—is a decreas- date, redshifts have been measured for just over a mil-
ing function of the frequency of observation: Sv ∝ v–α. The lion sources; and even the largest surveys such as the
number of sources above some limiting flux density, per Sloan Digital Sky Survey, cover a relatively small portion
unit solid-angle Ω, is also a decreasing function of the of the sky. The Ellis and Baldwin test requires a similar
threshold flux dN/dΩ(> Sv) ∝ Sv–x. The net effect is that number of sources spread uniformly on the sky. Fortu-

4 / 10
 INFERENCE / Vol. 6, No. 4

nately such a catalog has become available recently from calibrated in a joint light-curve analysis.37 The measured
the Wide-Field Infrared Survey Explorer (WISE) satellite. redshifts and magnitudes had all been boosted to the
These data were used by my collaborator Nathan Secrest cosmic microwave background frame. Corrections were
and his coauthors to construct a uniform catalog of 1.36 made for the peculiar velocities of the supernova host gal-
million active galactic nuclei.32 With our collaborators axies in our bulk flow, assuming that there is convergence
Jacques Colin, Sebastian von Hausegger, Roya Mohayaee, to the cosmic microwave background frame beyond ~150
and Mohamed Rameez, we applied various quality cuts Mpc.
on the CatWISE2020 catalog and masked out problem- Motivated by the work of Christos Tsagas, we had rea-
atic regions, including a band of ± 30º around the galactic sons to question this analysis.38 Tsagas had observed that
plane.33 This revealed a strong dipole signature, its direc- tilted observers embedded in a bulk flow may erroneously
tion consistent with the cosmic microwave background conclude that they are accelerating even when the expan-
dipole. But the amplitude of 𝒟 = 0.0154 was over twice sion rate is globally decelerating. A clear signature would
as large as the expected value of 0.0072 using the mean be a dipole asymmetry in the derived q0 toward the bulk-
values of ⟨α⟩ ~ 1.26 and ⟨x⟩ ~ 1.7 appropriate for this cat- flow direction.39 In the ΛCDM model, the SNe Ia data is
alog. We thus confirmed the anomaly first revealed by the fitted to the magnitude-redshift relationship in Equation
NRAO VLA Sky Survey catalog. A representative subset of 7. On the other hand, to determine a kinematic quantity
these quasars have spectroscopic redshift measurements like acceleration, the luminosity distance can be expanded
in the Sloan Digital Sky Survey. Their mean redshift is in terms of increasing derivatives of H0, q0, and the jerk
⟨z⟩ = 1.2. Fewer than 1% of the sources are at z < 0.1. parameter j0 = ä/aH 3| 0. The result is a power series that is
The anomalously large dipole is not of local origin. adequately accurate for z ≲ 1:

T
o quantify this discrepancy, we simulated ten
cz 1
million mock skies by randomly sampling the 10. dL = 1 + (1 – q0)z
quasar catalog according to the distributions of H0 2
flux densities and spectral indices. We applied the same
masks and flux cuts as for the real sky, and then intro-
+
1
6( kc2
)
1 – q0 – 3q02 + j0+ 2 2 z2 + … .
H0 a0
duced relativistic aberration and Doppler boosting with
β = 0.00123 as per the kinematic interpretation of the
cosmic microwave background dipole. Such was our In order to be model-independent, we used the above
null hypothesis. The calculation yielded a distribution of equation, but allowed the deceleration parameter to have
expectations for the quasar dipole. Only 5 out of 10 million both direction and scale dependence:
simulated skies exhibit a dipole as large as the real one.
The null hypothesis was rejected with a p-value of 5 × 10–7, 11. q = qm + qd ⋅ 𝑛̂ ℱ(z, S),
which corresponds to a significance of 4.9σ for the normal
distribution. The CatWISE2020 catalog has almost no where qm and qd are the monopole and dipole components,
objects in common with the NRAO VLA Sky Survey cata- 𝑛̂ is the direction of the dipole axis, and ℱ(z, S) describes
log. Combining their results forces us to reject the standard its scale dependence. The best fit was found to be an expo-
cosmological model expectation at nearly 6σ.34 nential form: exp(–z, S). To determine the best fit to the
This anomaly can no longer be dismissed. It appears joint light-curve analysis data, we used the statistically
that the cosmic rest frames of matter traced by quasars principled maximum likelihood estimator we had earlier
and the cosmic microwave background do not coincide. constructed,40 rather than adjusting error bars to fit the
This strikes at the very heart of the FLRW models, as ΛCDM model.41 This revealed that the inferred acceler-
Ellis and Baldwin noted.35 And it calls into question all the ation is indeed a dipole and is ~50 times larger than the
inferences drawn from analysis of cosmological data in monopole. It falls off exponentially with a scale length of
this framework: in particular, the inference that the uni- ~80/h Mpc. The monopole qm is consistent with being
verse is dominated by a cosmological constant. zero at just 1.4σ. But the dipole component qd is not zero
Still, none of these probes is directly sensitive to Λ. Since at 3.9σ.42 This is in accordance with Tsagas’s expectations
the value of Λ is inferred from the cosmic sum rule, their from the Raychaudhuri equation of general relativity and
analysis relies on the assumptions of isotropy and homo- clearly demonstrates that the inferred acceleration is not
geneity. There is no independent evidence of accelerated due to a cosmological constant.43
expansion from any of the low redshift probes. They are It exists because we are non-Copernican observers
equally consistent with expansion at a constant rate.36 embedded in a deep bulk flow.44
In 2014, a collaboration of nearly all of the world’s The original analyses by the Supernova Cosmology Proj-
supernova astronomers made public a catalog of 740 ect and High-z Supernova Search teams had employed,
spectroscopically confirmed Type Ia (SNe Ia) superno- respectively, just 60 and 50 SNe Ia, 17 of which were in
vae compiled from all available surveys and uniformly common. We found that most of these SNe Ia were also in

5 / 10
 CRITICAL ESSAYS

the general direction where qd peaks.45 Forthcoming data Taking into account a large body of work besides the Geller,
from the Dark Energy Spectroscopic Instrument and the de Lapparent, Huchra work—your own work on the large-
Rubin Observatory Legacy Survey of Space and Time will scale motions and the work of the Seven Samurai and all
establish definitively whether the inferred cosmic acceler- of that work which has shown that the universe is more
ation is indeed anisotropic. If so, dark energy will be ruled inhomogeneous than might have been present in simple
out as the explanation. models—has that altered your view of the big bang model
In FLRW cosmology, the value assigned to Λ is inferred at all, or of the validity of the model, the assumptions of the
from the cosmic sum rule, which itself follows from the model, that kind of thing?
assumptions of homogeneity and isotropy encapsulated
in the FLRW metric. More complicated metrics can be To this her reply was,
formulated for inhomogeneous and anisotropic uni-
verses,46 but such models have not been explored in as It certainly has convinced me that we’re not living in a
much detail as the FLRW models, especially with regard homogeneous, isotropic [universe]. I mean these things
to structure formation. In such extended models, Ein- that I really suspected in the back of my mind, I can now
stein’s equations do not simplify to a straightforward sum say publicly. I’m not sure the Robertson–Walker universe
rule. Additional terms are required and the inference of Λ exists. … If someone came out with a different model that
becomes ambiguous. The Lemaître–Tolman–Bondi model could incorporate such large-scale inhomogeneities, I
provides an exact solution of Einstein’s equations for a would be delighted to see it, but until then I will just live
radially inhomogeneous but isotropic universe. The same with the big bang model.52
data can be fitted by an appropriate radial variation of the
metric, rather than with Λ.47 Though we see gravitational We do not yet have the definitive different model, but
instability develop from the small density fluctuations we hope to have further motivated crucial tests of the cur-
imprinted on the cosmic microwave background, the evo- rent model with forthcoming data, which will provide the
lution of cosmic structure in such a model has yet to be clues necessary to formulate it.53 We believe that the dis-
fully explored. By contrast, the FLRW model successfully covery of dark energy was an accident waiting to happen
describes the entire evolution of the universe. Using the given the oversimplified framework of the standard model
fundamental laws encoded in the standard model of parti- of cosmology.
cle physics, we can in fact extrapolate reliably as far back
as ~10–12 seconds after the Big Bang.48 Subir Sarkar is Emeritus Professor at the Rudolf Peierls
The universe becomes simpler as we go back to such Centre for Theoretical Physics, University of Oxford.
early epochs. The energy density becomes dominated by
matter, and then by radiation, which scales even faster
with the redshift, when Ωm(1 + z)3 is overtaken by Ωr(1 + z)4
at z ≳ 104. The energy density is then well-described by
the Einstein–de Sitter model, which has zero curvature
and zero Λ. It is only in the late universe at z ≲ 1 that this 1. Although this snide put-down of cosmologists by Landau is
oversimplified framework leads us to infer domination often quoted, I have been unable to find a reliable source for
by unphysical dark energy. It is here that the cosmologi- it. Somewhat disturbingly, Oxford Essential Quotations, ed.
cal model must be made more sophisticated to take into Susan Ratcliffe (Oxford, UK: Oxford University Press, 2018),
account the inhomogeneities that arise due to gravitational cites the attribution as originating in a popular book, Big
collapse. Bang (2004) by Simon Singh.
Major attempts along these lines are the backreaction 2. Throughout this essay, the standard model always refers
program led by Thomas Buchert and the ambitious times- to the standard model of cosmology, and not the Standard
cape cosmology of David Wiltshire.49 These proposals are Model of particle physics.
controversial and have not been widely accepted.50 Per- 3. In Paul Arthur, Albert Einstein: Philosopher-Scientist (Evan-
haps that is simply because the ΛCDM model is so much ston, IL: Library of Living Philosophers, 1949), 443.
simpler and easier to confront with a wide variety of data 4. In 1930, Arthur Eddington demonstrated that Einstein’s
such as gravitational lensing, baryon acoustic oscillations, static universe was unstable.
galaxy clusters, and fluctuations in the cosmic microwave 5. Albert Einstein, “Cosmological Considerations in the Gen-
background. All are supposedly concordant in this frame- eral Theory of Relativity,” Sitzungsberichte der Preussischen
work.51 Akademie der Wissenschaften (Berlin) 142 (1917), trans.

I
W. Perret and G. B. Jeffrey, in Hendrick Lorentz et al., The
t is not new, this realization that the universe we Principle of Relativity (New York: Dover, 1852), 175–88; and
inhabit does not look like the idealized FLRW model. Willem de Sitter, “On Einstein’s Theory of Gravitation and
Rubin, when interviewed for an oral history project in Its Astronomical Consequences,” Monthly Notes of the Royal
1998, was asked, Astronomical Society 78, no. 3 (1917).

6 / 10
 INFERENCE / Vol. 6, No. 4

Alexander Friedmann, “Über die Krümmung des Raumes problem of Λ in quantum field theory cannot be solved by
(On the Curvature of Space),” Zeitschrift für Physik 10 (1922): invoking a new underlying symmetry, at least not in 3 + 1
377–86, doi:10.1007/BF01332580; Alexander Friedmann, dimensions.
“Über die Möglichkeit einer Welt mit konstanter negati- 9. It is supplemented by a second equation for the rate of
ver Krümmung des Raumes (On the Possibility of a World change of the expansion rate:
with Constant Negative Curvature of Space),” Zeitschrift für
Physik 21 (1924): 326–32, doi:10.1007/BF01328280; Georges ä/a = (4πGN / 3)(𝜌 + 3p),
Lemaître, “Un Univers homogène de masse constante et
de rayon croissant rendant compte de la vitesse radiale des which shows that the Hubble parameter decreases with
nébuleuses extra-galactiques (A Homogeneous Universe of time unless (𝜌 + 3p) is negative: p < –𝜌/3. Ordinary matter
Constant Mass and Increasing Radius Accounting for the is effectively pressureless. In the case of radiation, which
Radial Velocity of Extra-Galactic Nebulae),” Annales de la dominated nonrelativistic matter in the hot early universe,
Société Scientifique de Bruxelles A47 (1927): 49–59; Georges p = 𝜌/3. In both cases, the expansion rate ought therefore
Lemaître, “L’Univers en expansion (The Expanding Uni- to slow down. Whence the definition of the deceleration
verse),” Annales de la Société Scientifique de Bruxelles A53 parameter as
(1933): 51–85; and Georges Lemaître, “The Beginning of the
World from the Point of View of Quantum Theory,” Nature q0 ≡ –äa / ȧ2.
127 (1931): 706, doi:10.1038/127706b0.
6. Contrary to the popular myth that it required adding unac- Given the cosmological constant p = –𝜌, so it can cause the
ceptably large numbers of epicycles to correct errors, the expansion rate to speed up if it comes to dominate over
geocentric system was never in such crisis. See Owen Gin- matter, thus reversing the sign of the deceleration param-
gerich, The Eye of Heaven: Ptolemy, Copernicus, Kepler (New eter:
York: American Institute of Physics, 1993). The heliocentric
theory prevailed not because it better fit the data, at least q0 = Ωm – ΩΛ / 2.
initially, but because it was a better theory, providing a phys-
ical basis for planetary motions. The increasing precision 10. David Branch and Gustav Tammann, “Type Ia Superno-
of the data eventually made a difference too. Unfortunately vae as Standard Candles,” Annual Review of Astronomy
there had been no data available when, sixteen centuries and Astrophysics 30 (1992): 359–89, doi:10.1146/annurev.
before Johannes Kepler, “Aristarchus of Samos brought out aa.30.090192.002043.
a book consisting of certain hypotheses … that the Earth 11. The supernova of 1006 CE, which went off only ~2.2 kpc
revolves about the Sun in the circumference of a circle, the away in the Milky Way, was bright enough to be visible in the
Sun lying in the middle of the orbit.” These words belong to daytime for several months. It features in numerous records
Archimedes of Syracuse, as noted in “The Sand Reckoner,” in the Far East and in Arab lands—yet was not acknowledged
in The Works of Archimedes: Edited in Modern Notation in Europe, except by the monk Hepidannus in the Abbey of
with Introductory Chapters, ed. Thomas Heath (Cambridge: St. Gallen in Switzerland, presumably because the heavens
Cambridge University Press, 2009), 221–32, doi:10.1017/ had been declared to be immutable by Aristotle and Ptolemy!
CBO9780511695124.017. It was not until 1572, when another such explosion hap-
7. For an accessible introduction to curved space-time in Ein- pened nearby, that Tycho Brahe recorded the supernova now
stein’s general theory of relativity, see Robert Osserman, named after him and measured its position more accurately
Poetry of the Universe (New York: Anchor Books, 1995). Do than had been done hitherto. In 1604, another supernova was
not be misled by the title—the author was chair of mathe- recorded by and is named after his student Kepler. None have
matics at Stanford University. But he manages somehow to been seen in the Milky Way since then—so one appears to be
avoid formulae while brilliantly communicating the under- overdue. See David Clark and Francis Stephenson, The His-
lying physical concepts. torical Supernovae (Oxford: Pergamon Press, 1977).
8. It is often said that Einstein added the lambda term to his 12. Neta Bahcall et al., “The Cosmic Triangle: Revealing the
equation to make the universe static, as it was believed to be State of the Universe,” Science 284, no. 5,419 (1999): 1,481–88,
the case at the time, but later recanted on learning that the doi:10.1126/science.284.5419.1481.
universe is in fact expanding and called it “his greatest blun- 13. Michael Hobson, George Efstathiou, and Anthony Lasenby,
der.” Apart from the fact that there is no factual evidence for General Relativity: An Introduction for Physicists (Cam-
the anecdote, which is due to George Gamow, it was never bridge, UK: Cambridge University Press, 2005), 187.
Einstein’s choice whether or not to add the term since the 14. Wolfgang Pauli, “Die allgemeinen Prinzipien der Wellen-
underlying symmetries of the field equations permit it to mechanik (The general principles of wave mechanics),”
be there. See Steven Weinberg, “The Cosmological Con- Handbuch der Physik 24 (1933). To appreciate this comment
stant Problem,” Reviews of Modern Physics 61, no. 1 (1989): in its appropriate historical context, see Norbert Straumann,
1–24, doi:10.1103/RevModPhys.61.1. In this classic review, “Dark Energy,” Approaches to Fundamental Physics 721
Weinberg presented a no-go theorem that the fine-tuning (2007): 327–97, doi:10.1007/978-3-540-71117-9_13.

7 / 10
 CRITICAL ESSAYS

15. Steven Weinberg quoted in James Glanz, “What Do Physi- and selection functions assuming the model they are sup-
cists Fret About? Nothing,” New York Times, November 30, posedly testing, apparently without even being aware that
1999. this is unprincipled. Such confirmation bias may account
16. Edward Arthur Milne, “World-Structure and the Expansion for the observation by Rupert Croft and Matthew Dailey,
of the Universe,” Zeitschrift für Astrophysik 6 (1933): 1–95. “On the Measurement of Cosmological Parameters,” arX-
This was a natural extension of the Copernican principle, iv:1112.3108v2 (2015), who noted that of the twenty-eight
which had earlier led to the overthrow of the heliocentric measurements of ΩΛ in our sample published since 2003,
universe model, and was implicitly assumed in the construc- only two are more than 1σ from the WMAP results. Wider
tion of the FLRW models. use of blind analyses in cosmology could help to avoid this.
Subsequently, Hermann Bondi and Tommy Gold If the above twenty-eight results had indeed been inde-
extended this further to the perfect cosmological princi- pendent, this number should have been typically nine, or
ple, whereby the universe does not occupy a special point 32% of the measurements.
in time either. Hermann Bondi and Thomas Gold, “The 18. J. M. Stewart and Dennis Sciama, “Peculiar Velocity of the
Steady-State Theory of the Expanding Universe,” Monthly Sun and Its Relation to the Cosmic Microwave Background,”
Notices of the Royal Astronomical Society 108, no. 3 (1948): Nature 216 (1967): 748, doi:10.1038/216748a0. See also Dennis
252–70, doi:10.1093/mnras/108.3.252. This idea, however Sciama, “Peculiar Velocity of the Sun and the Cosmic Micro-
elegant, was nonetheless observationally ruled out following wave Background,” Physical Review Letters 18, no. 24 (1967):
the discovery of the cosmic microwave background and the 1,065, doi:10.1103/PhysRevLett.18.1065. I had the great for-
realization that we did originate about 14 billion years ago in tune of being inspired to work in cosmology by Sciama, who
what Fred Hoyle, a key proponent of the steady-state theory, had mentored many key figures in the field including George
dismissively called “the Big Bang.” The name stuck, and so Ellis, Martin Rees, Stephen Hawking, and Roger Penrose. I
has the weaker form of the spatial cosmological principle— recall he used to lament that this paper, which he considered
until now when we finally have the data to test it. one of his best insights, has hardly ever been cited!
17. David Hogg et al., “Cosmic Homogeneity Demonstrated 19. Phillip Peebles and David Wilkinson, “Comment on the
with Luminous Red Galaxies,” Astrophysical Journal 624, no. Anisotropy of the Primeval Fireball,” Physical Review 174, no.
1 (2005): 54–58, doi:10.1086/429084. Note that this test was 5 (1968): 2,168, doi:10.1103/PhysRev.174.2168.
undertaken relatively recently—following the first large- 20. For a review of the experiments, see George Smoot, “Nobel
scale map of the galaxy distribution by the Sloan Digital Sky Lecture: Cosmic Microwave Background Radiation Anisot-
Survey, which began operations in 2000 after a decade of ropies: Their Discovery and Utilization,” Reviews of Modern
design and construction. There remain concerns about the Physics 79, no. 4 (2006): 1,349–80, doi:10.1103/RevMod-
validity of the result because such surveys have simply not Phys.79.1349. Fig. 4 shows the velocity components of the
been big enough to provide an unbiased result. For example, cosmic microwave background dipole.
in the often-cited WiggleZ Dark Energy Survey, which mea- 21. Nabila Aghanim et al., “Planck 2018 Results I. Overview and
sured redshifts for only 200,000 galaxies, the largest spheres the Cosmological Legacy of Planck,” Astronomy and Astro-
in which galaxies were being counted and claimed to be scal- physics 641 (2020): A1, doi:10.1051/0004-6361/201833880.
ing as r3 actually extended beyond the survey volume, so the 22. Vera Rubin, W. Kent Ford Jr., and Judith Rubin, “A Curi-
numbers in the unobserved regions were simulated assum- ous Distribution of Radial Velocities of ScI Galaxies with
ing a random (i.e., homogeneous) distribution. See Morag 14.0 ≤ m ≤ 15.0,” Astrophysical Journal Letters 183 (1973):
Scrimgeour et al., “The WiggleZ Dark Energy Survey: The L111, doi:10.1086/181265; Vera Rubin et al., “Motion of
Transition to Large-Scale Cosmic Homogeneity,” Monthly the Galaxy and the Local Group Determined from the
Notices of the Royal Astronomical Society 425, no. 1 (2012): Velocity Anisotropy of Distant Sc I Galaxies II. The Anal-
116–34, doi:10.1111/j.1365-2966.2012.21402.x. That is of ysis for the Motion,” Astronomical Journal 81, no. 9 (1976):
course inserting by the back door the answer being sought! 719–37, doi:10.1086/111943; and Vera Rubin, “Is There Evi-
This self-serving strategy has also been adopted in subse- dence for Anisotropy in the Hubble Expansion?,” Annals
quent work as a bias correction: e.g., Rafael Silva Gonçalves of the New York Academy of Sciences 302 (1977): 408–21,
et al., “Measuring the Scale of Cosmic Homogeneity with doi:10.1111/j.1749-6632.1977.tb37062.x.
SDSS-IV DR14 Quasars,” Monthly Notices of the Royal Astro- 23. Alan Dressler et al., “Spectroscopy and Photometry of
nomical Society 481, no. 4 (2018): 5,270–74, doi:10.1093/ Elliptical Galaxies: A Large-Scale Streaming Motion in the
mnras/sty2670. To make real progress will require much Local Universe,” Astrophysical Journal Letters 313 (1987):
bigger surveys, as with the Dark Energy Spectroscopic L37, doi:10.1086/184827; Donald Lynden-Bell et al., “Pho-
Instrument, which has begun measuring redshifts for sev- tometry and Spectroscopy of Elliptical Galaxies. V. Galaxy
eral million galaxies and quasars, and the forthcoming Rubin Streaming toward the New Supergalactic Center,” Astro-
Observatory Legacy Survey of Space and Time. physical Journal 326 (1988): 19, doi:10.1086/166066; and
As cosmological analyses have become increasingly com- Alan Dressler, “The Great Attractor: Do Galaxies Trace the
plex, researchers have adopted circular reasoning: they Large-Scale Mass Distribution?,” Nature 350 (1991): 391–97,
estimate the relevant parameters, measurement efficiencies, doi:10.1038/350391a0.

8 / 10
 INFERENCE / Vol. 6, No. 4

24. Michael Rowan-Robinson et al., “A Sparse-Sampled Red- the Royal Astronomical Society 471, no. 1 (2017): 1,045–55,
shift Survey of IRAS Galaxies. I. The Convergence of the doi:10.1093/mnras/stx1631.
IRAS Dipole and the Origin of Our Motion with Respect to 31. For a Gaussian distribution of probabilities, “3σ” implies
the Microwave Background,” Monthly Notices of the Royal odds of 1 in 4,000 that the observation is just a random fluc-
Astronomical Society 247 (1990): 1–18; Dale Kocevski and tuation. When there is no actual signal, it is called the null
Harald Ebeling, “On the Origin of the Local Group’s Pecu- hypothesis. For discoveries of fundamental significance, it
liar Velocity,” Astrophysical Journal 645, no. 2 (2006): 1,043, is usual to require at least 5σ significance, i.e., odds of 1 in
doi:10.1086/503666; Guilhem Lavaux et al., “Cosmic Flow 3.5 million. In practice, the relevant probability distribu-
from Two Micron All-Sky Redshift Survey: The Origin of tion function is hardly ever Gaussian in a real experiment,
Cosmic Microwave Background Dipole and Implications for especially at large excursions from the mean. Real experi-
ΛCDM Cosmology,” Astrophysical Journal 709, no. 1 (2010): ments typically have systematic, in addition to statistical,
483, doi:10.1088/0004-637X/709/1/483; and Hume Feld- uncertainties. Nevertheless, sigmas are still a popular way to
man, Richard Watkins, and Michael Hudson, “Cosmic Flows parameterize the degree of unlikelihood of the observation.
on 100 h–1 Mpc Scales: Standardized Minimum Variance Note that it is not the probability that the null hypothesis is
Bulk Flow, Shear and Octupole Moments,” Monthly Notices incorrect: the probability of the data given the model does
of the Royal Astronomical Society 407, no. 4 (2010): 2,328–38, not equal the probability of the model given the data! See
doi:10.1111/j.1365-2966.2010.17052.x. Evelyn Lamb, “5 Sigma What’s That?,” Scientific Ameri-
25. Jacques Colin et al., “Probing the Anisotropic Local Uni- can, July 17, 2012; and Louis Lyons, “Statistical Issues in
verse and Beyond with SNe Ia Data,” Monthly Notices of Particle Physics,” in Particle Physics Reference Library, ed.
the Royal Astronomical Society 414, no. 1 (2011): 264–71, C. Fabjan and Herwig Schopper (Cham: Springer, 2020),
doi:10.1111/j.1365-2966.2011.18402.x; and U. Feindt et al., doi:10.1007/978-3-030-35318-6_15.
“Measuring Cosmic Bulk Flows with Type Ia Superno- 32. Nathan Secrest et al., “Identification of 1.4 Million
vae from the Nearby Supernova Factory,” Astronomy and Active Galactic Nuclei in the Mid-Infrared Using WISE
Astrophysics 560 (2013): A90, doi:10.1051/0004-6361/ Data,” Astrophysical Journal Supplement 221 (2015): 12,
201321880. doi:10.1088/0067-0049/221/1/12.
26. R. Brent Tully et al., “The Laniakea Supercluster of Gal- 33. Nathan Secrest et al., “A Test of the Cosmological Princi-
axies,” Nature 513 (2014): 71–73, doi:10.1038/nature13674. ple with Quasars,” Astrophysical Journal Letters 908, L51
See Noam Libeskind and R. Brent Tully, “Our Place in the (2021), doi:10.3847/2041-8213/abdd40/. These are super-
Cosmos,” Scientific American 315 (2016): 32–39, doi:10.1038/ massive spinning black holes—millions to billions of solar
scientificamerican0716-32. masses—at the centers of galaxies, surrounded by an equa-
27. James Gunn, “Deviations from Pure Hubble Flow: A torial accretion disk of infalling matter. They shoot out huge
Review,” Astronomical Society of the Pacific Conference jets of plasma at relativistic speeds that create extended
Series 4 (1988): 357. radio sources when entrained by the intergalactic medium.
28. George Ellis and John Baldwin, “On the Expected Anisot- They were originally called quasi-stellar radio sources, or
ropy of Radio Source Counts,” Monthly Notices of the Royal quasars, since they appeared as distant pointlike sources.
Astronomical Society 206, no. 2 (1984): 377, doi:10.1093/ Today the host galaxy can usually be resolved as well. Some-
mnras/206.2.377. times by chance one of the plasma jets is aimed directly at
29. Ellis and Baldwin, “On the Expected Anisotropy of Radio us. The source is then particularly intense and is called a
Source Counts.” “blazar.”
30. Chris Blake and Jasper Wall, “A Velocity Dipole in the Dis- 34. This result is to be published.
tribution of Radio Galaxies,” Nature 416 (2002): 150–52, 35. Ellis and Baldwin, “On the Expected Anisotropy of Radio
doi:10.1038/416150a; Ashok Singal, “Large Peculiar Motion Source Counts.”
of the Solar System from the Dipole Anisotropy in Sky Bright- 36. I. Tutusaus et al., “Is Cosmic Acceleration Proven by Local
ness Due to Distant Radio Sources,” Astrophysical Journal Cosmological Probes?,” Astronomy and Astrophysics 602
742, no. 2 (2011): L23, doi:10.1088/2041-8205/742/2/L23; (2017): A73, doi:10.1051/0004-6361/201630289.
Cameron Gibelyou and Dragan Huterer, “Dipoles in the Sky,” 37. M. Betoule et al., “Improved Cosmological Constraints
Monthly Notices of the Royal Astronomical Society 427, no. 3 from a Joint Analysis of the SDSS-II and SNLS Supernova
(2012): 1,994–2,021, doi:10.1111/j.1365-2966.2012.22032.x; M. Samples,” Astronomy and Astrophysics 568 (2014): A22,
Rubart and D. J. Schwarz, “Cosmic Radio Dipole from NVSS doi:10.1051/0004-6361/201423413.
and WENSS,” Astronomy and Astrophysics 555 (2013): A117, 38. Jacques Colin et al., “Evidence for Anisotropy of Cosmic
doi:10.1051/0004-6361/201321215; Prabhakar Tiwari et al., Acceleration,” Astronomy and Astrophysics Letters 631
“Dipole Anisotropy in Sky Brightness and Source Count (2019): L13, doi:10.1051/0004-6361/201936373.
Distribution in Radio NVSS Data,” Astroparticle Physics 39. Christos Tsagas, “Peculiar Motions, Accelerated Expansion
61 (2015): 1–11, doi:10.1016/j.astropartphys.2014.06.004; and the Cosmological Axis,” Physical Review D 84 (2011):
and Jacques Colin et al., “High-Redshift Radio Galaxies 063503, doi:10.1103/PhysRevD.84.063503; and Christos
and Divergence from the CMB Dipole,” Monthly Notices of Tsagas and Miltiadis Kadiltzoglou, “Deceleration Param-

9 / 10
 CRITICAL ESSAYS

eter in Tilted Friedmann Universes,” Physical Review D 92 47. Marie-Noëlle Célérier, “Do We Really See a Cosmolog-
(2015): 043515, doi:10.1103/PhysRevD.92.043515. ical Constant in the Supernovae Data?,” Astronomy and
40. J. T. Nielsen, A. Guffanti, and Subir Sarkar, “Marginal Evi- Astrophysics 353 (2000): 63–71. This proposal has been
dence for Cosmic Acceleration from Type Ia Supernovae,” much misunderstood, as explained in her later paper:
Nature Scientific Reports 6, no. 35,596 (2016), doi:10.1038/ “Some Clarifications about Lemaître-Tolman Models of
srep35596. the Universe Used to Deal with the Dark Energy Prob-
41. Betoule et al., “Improved Cosmological Constraints.” lem,” Astronomy and Astrophysics 543 (2012): A71,
42. Colin et al., “Evidence for Anisotropy of Cosmic Accelera- doi:10.1051/0004-6361/201219104.
tion.” 48. Valerie Rubakov and Dmitry Gorbunov, Introduction to the
43. Tsagas, “Peculiar Motions, Accelerated Expansion and the Theory of the Early Universe (Singapore: World Scientific
Cosmological Axis”; and Tsagas and Kadiltzoglou, “Deceler- Press, 2018).
ation Parameter in Tilted Friedmann Universes.” 49. Thomas Buchert, “On Average Properties of Inho-
44. Our findings were reproduced by Daniel Rubin and Jessica mogeneous Fluids in General Relativity. 1. Dust
Heitlauf, “Is the Expansion of the Universe Accelerating? Cosmologies,” Classical and Quantum Gravity 32 (2000):
All Signs Still Point to Yes a Local Dipole Anisotropy Cannot 105–25, doi:10.1023/A:1001800617177; Thomas Buchert,
Explain Dark Energy,” Astrophysical Journal Letters 894 “Dark Energy from Structure: A Status Report,” Classi-
(2020): 68, doi:10.3847/1538-4357/ab7a16. Nevertheless, cal and Quantum Gravity 40 (2008): 467–527, doi:10.1007/
these authors assert that the dipole anisotropy we had found s10714-007-0554-8; David Wiltshire, “Cosmic Clocks,
in the deceleration parameter can be gotten rid of by trans- Cosmic Variance and Cosmic Averages,” New Journal of
forming to the cosmic microwave background frame and Physics 9 (2007): 377, doi:10.1088/1367-2630/9/10/377; and
making corrections for peculiar velocities of the SNe Ia host David Wiltshire, “Average Observational Quantities in the
galaxies. Moreover, they claim that the significance of the Timescape Cosmology,” Physical Review D 80, no. 12 (2009):
monopole term being nonzero can be raised by allowing for 123,512, doi:10.1103/PhysRevD.80.123512.
ad-hoc redshift- and sample-dependence of the stretch and 50. See, e.g., Akihiro Ishibashi and Robert Wald, “Can the
color corrections to the SNe Ia light curves, which requires Acceleration of Our Universe Be Explained by the Effects
more than doubling the number of parameters. We disagree of Inhomogeneities?,” Classical and Quantum Gravity 23,
that any of this is statistically justified. See Jacques Colin no. 1 (2006): 235, doi:10.1088/0264-9381/23/1/012; and
et al., “A Response to Rubin & Heitlauf: ‘Is the Expansion Thomas Buchert et al., “Is There Proof That Backreaction
of the Universe Accelerating? All Signs Still Point to Yes’,” of Inhomogeneities Is Irrelevant in Cosmology?,” Classical
arXiv:1912.04257 (2019). Their procedure is, in any case, and Quantum Gravity 32 (2015): 215,021, doi:10.1088/0264-
thoroughly undermined by our subsequent finding that cos- 9381/32/21/215021.
mologically distant sources are not isotropic in the cosmic 51. See, e.g., David Weinberg et al., “Observational Probes of
microwave background frame: Nathan Secrest et al., “A Test Cosmic Acceleration,” Physics Reports 530 (2013): 87–255,
of the Cosmological Principle with Quasars,” Astrophysical doi:10.1016/j.physrep.2013.05.001; and Dragan Huterer and
Journal Letters 908, no. 2 (2021): L51, doi:10.3847/2041- Daniel Shafer, “Dark Energy Two Decades After: Observables,
8213/abdd40. Probes, Consistency Tests,” Reports on Progress in Physics 81,
45. For a review, see Roya Mohayaee, Mohamed Rameez, and no. 1 (2018): 016901, doi:10.1088/1361-6633/aa997e.
Subir Sarkar, “Do Supernovae Indicate an Accelerating Uni- 52. Alan Lightman, interview with Vera Rubin, Washington,
verse?,” European Physical Journal Special Topics 230 (2021): D.C., April 3, 1989, American Institute of Physics: Oral His-
2,067–76, doi:10.1140/epjs/s11734-021-00199-6. tory Interviews.
46. See, e.g., Andrzej Krasiński, Inhomogeneous Cosmological 53. George Ellis and William Stoeger, “The ‘Fitting Problem’ in
Models (Cambridge, UK: Cambridge University Press, 1997). Cosmology,” Classical and Quantum Gravity 4, no. 6 (1987):
When I met Krasiński in Warsaw, I rashly joked that most 1,697, doi:10.1088/0264-9381/4/6/025.
cosmologists had never needed to read his book… I may have
spoken too soon! DOI: 10.37282/991819.22.21

Published on March 30, 2022

https://inference-review.com/article/heart-of-darkness

10 / 10