Does the new much-faster-speed-of-light theory fix the big bang’s problems?

A recent paper1 by Niayesh Afshordi and João Magueijo asserts that they have discovered a testable cosmology wherein during a “critical” cosmological phase of the early universe the maximal speed of propagation of matter (and hence light) was enormously much faster than the current speed of light (c) and faster than the speed of gravity, which in Einstein’s theory is the canonical speed c. They revisit what has become to be known as varying speed of light (VSL) models, in contrast to the now popular cosmic inflation models. They believe light travelled much faster just after the big bang than it does now and have developed a mathematical model of a big bang universe only a miniscule fraction of a second after the alleged hot beginning of the Universe.


João Magueijo at the journée de la Science at the EPFL, 11 November 2005. Credit: Wikipedia

The big bang model has many problems, but the biggest and most difficult to solve is what is known as the ‘horizon problem’.2 Cosmic inflation has been invoked to solve this problem. Afshordi and Magueijo agree that,

… the Big Bang model of the Universe remains an unfinished work of art. Many of its late-time successes can be traced to the initial conditions postulated for its early stages, and these are put in by hand, without justification, other than to retrofit the data. The main culprit for this shortcoming is the so-called horizon problem: the cosmological structures we observe today span scales that lay outside the ever-shrinking “horizons” of physical contact that plagued the early universe. This precludes a causal explanation for their initial conditions.1 (emphases added)

Cosmologist believe that structure in the universe was seeded from initial density variations in the early universe. But for structures (clusters of galaxies, for example) to naturalistically form gravity must propagate over the scale of any structure in the timescale available to it at the past epoch when the structures were allegedly built. In addition we observe a uniform temperature across all the sky in the cosmic microwave background (CMB) radiation, yet sources on opposite sides of the observable universe have not had time to exchange energy, at the constant speed of light c, in the time available in the big bang universe. That is, they have not had to time to come into thermal equilibrium. These limitations are what are known as ‘horizons’. The major problem with the big bang model is that cosmic inflation scenarios are inserted by hand, to overcome these ‘horizons’ but without any justification for why inflation started and why it stopped. Quite obviously if the speed of light were infinite there would exist no such ‘horizon’ to thermal equilibration of the Universe.

Various extensions to the big bang model have been proposed to overcome the horizon problems. Some proposals include an early accelerated expansion, a contracting phase followed by a bounce, a loitering early stage, and others a varying speed of light (VSL).

None of these proposals evades the criticism that retrofitting the data is still used to select in detail the primordial fluctuations that the model should produce. Once primordial causal contact is established, work can start on concrete physical mechanisms for spoiling perfect homogeneity (e.g. vacuum quantum fluctuations or thermal fluctuations). Typically it is found that one can produce a wide range of initial conditions including, but not circumscribed to those explaining the observations.1 (emphases added)

So the problem to date has been not what one would call prediction but retro- or post-diction. The result is known and the desired effect is put in by hand. Because cosmologists believe that the CMB radiation is leftover radiation from the early big bang universe, they also believe that the CMB temperature variations across various physical scales are representative of density variations in the early universe across those scales. By choosing a ΛCDM big bang cosmology3 they believe that they can ‘wind the clock back’ to some early stages of the big bang universe. (Note the circular reasoning in that alone.)

Typically they have used the precise measurements of temperature fluctuations (called anisotropies) in the CMB radiation to determine a power spectrum for those fluctuations—that quantifies the amplitude of density variations over various spatial scales. Those fluctuations can be quantified by a number called a spectral index nS from the power spectrum. The spectral index nS measurements within a certain range are claimed to indicate the validity of cosmic inflation.4 The range of inflation models limit the allowed value of this number.

Its value is interpreted to be related to early stage density fluctuations and as you might imagine the speed of sound in the early big bang universe is important to how those density fluctuations might have propagated. Therefore the speed of sound, which must be always less than the speed of light, is an important factor. If the speed of light is allowed to go to infinity then also the speed of sound can too. The Afshordi-Magueijo model involves a fast phase transition in the speed of sound, when the universe was very small, which leads to thermal fluctuations consistent with the measurement of nS from the CMB radiation.

Believing that the CMB does represent the oldest light in the universe, i.e. from the big bang fireball, Afshordi and Magueijo now also believe their theory can be tested using future observations of CMB radiation. They assert that light travelled at an infinite speed following the big bang, before slowing down to what we define as the speed of light, c, today (which is about 300,000 km/s or 186,000 miles/s). This much faster speed of light in the early universe then allegedly explains the uniform temperature of the CMB radiation seen in all directions around Earth, i.e. overcoming that ‘horizon’ problem, but not of the formation-of-structure ‘horizon’ (homogeneity and isotropy problems) because the speed of gravity is still c.

Their “critical cosmological model makes an unambiguous, non-tuned prediction for the spectral index of the scalar fluctuations: nS = 0.96478(64).”1 This value lies very much within the latest determination of the spectral index from the 2015 Planck satellite data5 (determined from CMB temperature fluctuations and an assumed cosmology) yet the authors are somewhat more circumspect.

The fact that its main prediction (for nS) lies spot in the middle of the Planck results should not beguile us into a false sense of security. Improved observations will soon vindicate or disprove this model.1

They could disprove their model but they cannot prove their model. This is the basis of science—disproof, never proof. Yet there still remains other problems that their proposal does not solve.

One may wonder about the status in our model of the other cosmological problems, such as the flatness, homogeneity and isotropy problems.1

The near infinite speed only pertains to massless matter (including light) but not gravity (gravitons travel at speed c). Except for flatness (the model requires exact flatness) the other cosmological problems are not solved by this new proposal. However the authors believe that their VSL model can solve them and deferred to a later paper in preparation.

However the main point in the new paper is that the authors claim their new model is testable.6 Operational science is based on the principle of repeatable testability. The question here is: Is their cosmology (or any cosmology) actually testable in an operational science sense?

The answer is an unequivocal no!

The reason for this is that the authors cannot interact with the Universe, which is what they really are wanting to test their model against. They may look for nullification of the spectral index against some future CMB data, which needs to improve in precision by at least 100 times. But even if it was found that some future improved result was tightly bound within their prediction, does it prove that the Universe had the history they claim their model describes?  No, it does not. This is because one cannot rule out all other models (including those not yet even dreamed up) that produce the same result. Cosmology in this case is not a repeatable science. I would even go so far as to say it is not really even science.7

The reason I say this is because even to generate the spectral index data one needs to assume a cosmology. Then there are a plethora of models, some of which may ‘predict’ the same value of certain parameters, most of which are derived this way.

I suggest, in fact, that the reason so many dark entities (dark matter, dark energy, dark radiation, dark photons, etc) are now suggested to get various theories to fit observational data is because the underlying theory – the big bang cosmology – is a bad theory for the structure and origin of the universe.8

The big bang never happened. The CMB radiation is real enough but to link it with a fictitious big bang fireball is a form of one’s belief system determining how to interpret the observational data.

Besides evidence exists to suggest that the CMB radiation is not even primordial (not the oldest light in the universe).9 Therefore if it is not from the big bang, how can any parameter related to its spectral density of temperature fluctuations be descriptive of the fictitious event?


  1. Afshordi, N., and Magueijo, J., Critical geometry of a thermal big bang, Physical Review D 94, 101301(R), November 18, 2016; preprint: arXiv:1603.03312v2.
  2. Lisle, J., Light-travel time: a problem for the big bang, Creation 25(4):48–49, September 2003.
  3. ΛCDM = dark energy and cold dark matter.
  4. Hartnett, J.G., Cosmic inflation: Did it really happen?, September 11, 2016.
  5. Ade, P.A.R., et al. (Planck), Planck 2015 results. XIII. Cosmological parameters, preprint: arXiv:1502.01589, 2015.
  6. Dunning, H., Theory that challenges Einstein’s physics could soon be put to the test,, November 25, 2016.
  7. Hartnett, J.G., Cosmology is not science!, December 28, 2013.
  8. Hartnett, J.G., Where materialism logically leads, May 31, 2016.
  9. Hartnett, J.G., ‘Light from the big bang’ casts no shadows, Creation 37(1):50-51, 2015.

Recommended Reading