3 Questions: Alan Guth on new insights into the ‘Big Bang’

Alan Guth

MIT physicist explains how new results bolster his 1980 theory of cosmic inflation.

Press Contact

Kimberly Allen
Email: allenkc@mit.edu
Phone: 617-253-2702
MIT News Office

Earlier this week, scientists announced that a telescope observing faint echoes of the so-called “Big Bang” had found evidence of the universe’s nearly instantaneous expansion from a mere dot into a dense ball containing more than 1090 particles. This discovery, using the BICEP2 telescope at the South Pole, provides the first strong evidence of “cosmic inflation” at the birth of our universe, when it expanded billions of times over. 

The theory of cosmic inflation was first proposed in 1980 by Alan Guth, now the Victor F. Weisskopf Professor of Physics at MIT. He discussed the significance of the new BICEP2 results with MIT News.

Q: Can you explain the theory of cosmic inflation that you first put forth in 1980?

A: I usually describe inflation as a theory of the “bang” of the Big Bang: It describes the propulsion mechanism that drove the universe into the period of tremendous expansion that we call the Big Bang. In its original form, the Big Bang theory never was a theory of the bang. It said nothing about what banged, why it banged, or what happened before it banged.

The original Big Bang theory was really a theory of the aftermath of the bang. The universe was already hot and dense, and already expanding at a fantastic rate. The theory described how the universe was cooled by the expansion, and how the expansion was slowed by the attractive force of gravity.

Inflation proposes that the expansion of the universe was driven by a repulsive form of gravity. According to Newton, gravity is a purely attractive force, but this changed with Einstein and the discovery of general relativity. General relativity describes gravity as a distortion of spacetime, and allows for the possibility of repulsive gravity.

Modern particle theories strongly suggest that at very high energies, there should exist forms of matter that create repulsive gravity. Inflation, in turn, proposes that at least a very small patch of the early universe was filled with this repulsive-gravity material. The initial patch could have been incredibly small, perhaps as small as 10-24 centimeter, about 100 billion times smaller than a single proton. The small patch would then start to exponentially expand under the influence of the repulsive gravity, doubling in size approximately every 10-37 second. To successfully describe our visible universe, the region would need to undergo at least 80 doublings, increasing its size to about 1 centimeter. It could have undergone significantly more doublings, but at least this number is needed.

During the period of exponential expansion, any ordinary material would thin out, with the density diminishing to almost nothing. The behavior in this case, however, is very different: The repulsive-gravity material actually maintains a constant density as it expands, no matter how much it expands! While this appears to be a blatant violation of the principle of the conservation of energy, it is actually perfectly consistent.

This loophole hinges on a peculiar feature of gravity: The energy of a gravitational field is negative. As the patch expands at constant density, more and more energy, in the form of matter, is created. But at the same time, more and more negative energy appears in the form of the gravitational field that is filling the region. The total energy remains constant, as it must, and therefore remains very small.

It is possible that the total energy of the entire universe is exactly zero, with the positive energy of matter completely canceled by the negative energy of gravity. I often say that the universe is the ultimate free lunch, since it actually requires no energy to produce a universe.

At some point the inflation ends because the repulsive-gravity material becomes metastable. The repulsive-gravity material decays into ordinary particles, producing a very hot soup of particles that form the starting point of the conventional Big Bang. At this point the repulsive gravity turns off, but the region continues to expand in a coasting pattern for billions of years to come. Thus, inflation is a prequel to the era that cosmologists call the Big Bang, although it of course occurred after the origin of the universe, which is often also called the Big Bang.

Q: What is the new result announced this week, and how does it provide critical support for your theory?

A: The stretching effect caused by the fantastic expansion of inflation tends to smooth things out — which is great for cosmology, because an ordinary explosion would presumably have left the universe very splotchy and irregular. The early universe, as we can see from the afterglow of the cosmic microwave background (CMB) radiation, was incredibly uniform, with a mass density that was constant to about one part in 100,000.

The tiny nonuniformities that did exist were then amplified by gravity: In places where the mass density was slightly higher than average, a stronger-than-average gravitational field was created, which pulled in still more matter, creating a yet stronger gravitational field. But to have structure form at all, there needed to be small nonuniformities at the end of inflation.

In inflationary models, these nonuniformities — which later produce stars, galaxies, and all the structure of the universe — are attributed to quantum theory. Quantum field theory implies that, on very short distance scales, everything is in a state of constant agitation. If we observed empty space with a hypothetical, and powerful, magnifying glass, we would see the electric and magnetic fields undergoing wild oscillations, with even electrons and positrons popping out of the vacuum and then rapidly disappearing. The effect of inflation, with its fantastic expansion, is to stretch these quantum fluctuations to macroscopic proportions.

The temperature nonuniformities in the cosmic microwave background were first measured in 1992 by the COBE satellite, and have since been measured with greater and greater precision by a long and spectacular series of ground-based, balloon-based, and satellite experiments. They have agreed very well with the predictions of inflation. These results, however, have not generally been seen as proof of inflation, in part because it is not clear that inflation is the only possible way that these fluctuations could have been produced.

The stretching effect of inflation, however, also acts on the geometry of space itself, which according to general relativity is flexible. Space can be compressed, stretched, or even twisted. The geometry of space also fluctuates on small scales, due to the physics of quantum theory, and inflation also stretches these fluctuations, producing gravity waves in the early universe.

The new result, by John Kovac and the BICEP2 collaboration, is a measurement of these gravity waves, at a very high level of confidence. They do not see the gravity waves directly, but instead they have constructed a very detailed map of the polarization of the CMB in a patch of the sky. They have observed a swirling pattern in the polarization (called “B modes”) that can be created only by gravity waves in the early universe, or by the gravitational lensing effect of matter in the late universe.

But the primordial gravity waves can be separated, because they tend to be on larger angular scales, so the BICEP2 team has decisively isolated their contribution. This is the first time that even a hint of these primordial gravity waves has been detected, and it is also the first time that any quantum properties of gravity have been directly observed.

Q: How would you describe the significance of these new findings, and your reaction to them?

A: The significance of these new findings is enormous. First of all, they help tremendously in confirming the picture of inflation. As far as we know, there is nothing other than inflation that can produce these gravity waves. Second, it tells us a lot about the details of inflation that we did not already know. In particular, it determines the energy density of the universe at the time of inflation, which is something that previously had a wide range of possibilities.

By determining the energy density of the universe at the time of inflation, the new result also tells us a lot about which detailed versions of inflation are still viable, and which are no longer viable. The current result is not by itself conclusive, but it points in the direction of the very simplest inflationary models that can be constructed.

Finally, and perhaps most importantly, the new result is not the final story, but is more like the opening of a new window. Now that these B modes have been found, the BICEP2 collaboration and many other groups will continue to study them. They provide a new tool to study the behavior of the early universe, including the process of inflation.

When I (and others) started working on the effect of quantum fluctuations in the early 1980s, I never thought that anybody would ever be able to measure these effects. To me it was really just a game, to see if my colleagues and I could agree on what the fluctuations would theoretically look like. So I am just astounded by the progress that astronomers have made in measuring these minute effects, and particularly by the new result of the BICEP2 team. Like all experimental results, we should wait for it to be confirmed by other groups before taking it as truth, but the group seems to have been very careful, and the result is very clean, so I think it is very likely that it will hold up.

Topics: Big Bang, 3 Questions, Astrophysics, Physics, cosmology, Quantum physics, cosmic inflation


Many great discoveries seem to raise just as many questions as they answer. Could Mr. Guth cite some major questions that have arisen from the data generated? For example, he speaks of "repulsive gravity material" in the interview. Do the data confirm any properties of this material?

djn6u observation/comment above was absolutely right. Like with Mr. Alan Guth, there's a lot of theory on the field but nothing on them is reliable/dependable in a series of questions that will be put forward. And to cut the story short, this is the hard part or fact for the great and erudite thinker to concede that in fact there's a consistent model of the origin and operation of the universe as one reader report said; though bigots in science and religion divide will not like it: Dennis Flores is in fact telling the truth throughout. Scholar however unknown on this field, the author of All But the World is Loving (ISBN 9781434983824 and upcoming Quintessence: the theory of everything - all but the world is loving 2)fulfilled Isaac newton theory on universal gravitation, big bang, black hole and some of its great advantage is that uncovering the moon quadruple motions couple with the earth and sun operation was in fact can predict if not precisely identify earthquake and tsunami.

I understand what you are saying, theory is not reality. There is a reason for theorectical assumptions and to check them against reality. But fluctuations in space are a reality. The thing is they come from somewhere. Most likely from the electro dynamic fields whih in reality would be attributed to a state that in principle can not attain equilibrium.

if you are talking something you can't prove, is like taking something against yourself. Electromagnetic wave produce by body such as sun and galaxy and somehow a massive planet or asteroid upon orbit creates wave like a bus or aircraft passing along. Measuring or taking data of a body frequency without the practical purpose is nonsense. I mean is that earth was in fact bombarded and or engulf with the solar frequency and that upon identifying this freq. could predict earthquake and tsunami and that could save life. Now which frequency the scientist had been detected, was it the one coming from the galaxy or from the sun which is the closest governing body in our dynamic but local [solar] system? It's depressing, until today astrophysicist were unaware of the true fundamental of the universe. No kidding, existing theory of black hole is absurd, where in with the help of a cup of coffee and a creamier, is in fact comprehensible.

The “B modes” polarization pattern is purported to represent the primordial gravity waves. However, the quantum fluctuations amplified by the inflation correspond to repulsive-gravity waves generated by repulsive-gravity material. As this repulsive-gravity material decays into an attractive-gravity one, attractive-gravity waves substitute for the repulsive-gravity ones, and therefore there is a moment during this transition where the value of gravity is zero and consequently space-time is flat. This flatness would erase any prior fluctuations in the fabric of space-time and primordial gravity waves would be concealed from observation forever. The observed pattern of polarization would correspond to an electromagnetic field other than the primordial gravitational.

According to the theory of inflation, the material that inflated at the origin of the universe was of a repulsive-gravity character, and presented quantum fluctuations in its density distribution.
The Einstenian analogy for gravity, as the deformation of an elastic flat surface - representing space-time - under the influence of a mass, would have this early universe represented by a series of minuscule bumps in the upwardly direction.

As the repulsive-gravity material was inflating - and these bumps were amplifying their size - it was also decaying into an attractive-gravity one, with the same density distribution and its corresponding downwardly bumps.

Every pair of these upwardly and downwardly bumps would cancel each other out - giving rise to a flat universe - at the moment where the quantity of attractive-gravity material was equal to the quantity of repulsive-gravity one.
From that moment on, the dominance of the attractive-gravity material would produce increasigly-growing downwardly bumps until the end of the decaying process.

This variation in the shape of space-time - from flatness to the downwardly bumpiness - would send off gravitational waves, whose electromagnetic field might correspond to the one indirectly observed in the "B modes" polarization pattern.

Albert Einstein gave us a new law of physics, c = 300,000 km/sec, and saw gravity as an acceleration, whereas Newton saw it as a force.

His field equations for the case of weak gravitational fields - such as those far away from big masses - show an entity, the gravitational waves, similar in form and speed to Maxwell’s electromagnetic waves for the case of absence of charges and currents, although dealing with compression and stretching of spacetime, instead of with electric and magnetic fields.

For strong gravitational fields, such as those in the proximity of a black hole or at the end of the hypothesized inflation preceding the Big Bang, the linearization of Einstein’s field equations is no longer possible, and its solution is far more complex than that for the weak gravitational fields.

In a first approximation, and considering the case of the weak gravitational fields, the cyclic orthogonal compression and stretching of spacetime, due to the gravitational waves generated at the time of the inflation, would have impinged on the cyclic orthogonal electric and magnetic fields of the first free photons, which are supposed to have appeared 380,000 years afterward.

This impingement is supposed to have consisted in the “B-modes” swirling polarization pattern of their electromagnetic field - today’s Cosmic Microwave Background radiation - as observed at a patch of the South Pole’s sky.

It would be interesting to know, for comparison purposes, what kind of impingement happens to occur in other - closer to us and of a lesser size - sources of both simultaneous and strong electromagnetic and gravitational emissions, such as the coalescing binary stars or black holes and the supernovae. This would be a step ahead in the identification and more precise knowledge of the gravitational waves, generally.

The theory of the Big Bang is based on the assumption that the red shifts of the electromagnetic spectra of the heavenly bodies follow the Hubble’s law, being linearly proportional to the receding velocities of the bodies in an expanding space-time, that is to say, mimicking the Doppler Effect.

There is, however, other type of red shifting, the gravitational or Einstein’s one, which is due to the fact that time on the surface of every one of the myriad of heavenly bodies in the universe is different from each other, and depends on the mass divided by the radius of the body in question.

If we take for instance a quasar, is its great red shift due to a great velocity of recession at a great distance from us, or is it due to an enormous mass divided by a relatively small radius at a shorter distance from us and at a slower velocity?

The same question may be posited with regard to the Cosmic Microwave Background. Is it due to the expansion of the wavelengths of the primordial photons of the Big Bang, or is it due to the red shifting - in differing degrees – of the whole myriad of radiant bodies in the universe?

Albert Einstein, in his book “RELATIVITY The Special and the General Theory” (Routledge, 1993, ISBN 0-415-09104-7), expresses (page 130) the following idea: “An atom absorbs or emits light of a frequency which is dependent on the potential of the gravitational field in which it is situated.” …...”Thus a displacement towards the red ought to take place for spectral lines produced at the surface of stars as compared with the spectral lines of the same element produced at the surface
of the earth, the amount of this displacement being

(νo –ν) /νo = (K*M) / (c squared *r) “ (where: νo, frequency of light at the surface of the earth; ν, frequency of light at the surface of the star; K, Newton´s constant of gravitation; M, mass of the star; c, velocity of light; r, radius of the star)

…...“For the sun, the displacement towards the red predicted by theory amounts to about two millionths of the wave-length. A trustworthy calculation is not possible in the case of the stars, because in general neither the mass M nor the radius r are known.”….”NOTE. - The displacement of spectral lines towards the red end of the spectrum was definitely established by Adams in 1924, by observations on the dense, companion of Sirius, for which the effect is about thirty times greater than for the sun. R.W.L.” (Robert W. Lawson, translator).

On the other hand, in “THE MEANING OF RELATIVITY” (Chapman & Hall, 1991, ISBN 0-412-20560-2) Albert Einstein elaborates (page 104) on the conclusive demonstration of the existence of the red shift of the spectral lines: ”This demonstration was made possible by the discovery of so-called ‘dwarf stars’ whose average density exceeds that of water by a factor of 10 to the power of 4. For
such a star (e.g. the faint companion of Sirius), whose mass and radius can be determined,* this red shift was expected, by the theory, to be about twenty times as large as for the sun, and indeed it was demonstrated to be within the expected range.* The mass is derived from the reaction on Sirius by spectroscopic means, using the Newtonian laws; the radius is derived from the total lightness and from the intensity of radiation per unit area , which may be derived from the temperature of its radiation.”

There is therefore a discrepancy in the theoretical expected value for the red shift: About thirty times greater than for the sun in the first book and about twenty times in the second.

This could perhaps be explained by an error in the above formula for the red shift.

Einstein’s formula for the frequency of light (page 130 of the first book)

ν = νo * [1 - ( ω squared * r squared) / ( c squared * 2 )] (where: ω, angular velocity of rotation; r, distance to the center of rotation)

It gives a different result for a unit mass on a rotating disc than in a Newtonian field.

On a rotating disc: - (ω squared * r squared) / 2 = Φc (where: Φc, the centripetal potential)

In a Newtonian field: - (ω squared * r squared) = - (K*M) / r = Φn (where: Φn, the
Newtonian potential)

Therefore the formula for the red shift in a Newtonian gravitational field would be

(νo –ν) /νo = [(ω squared * r squared) / (c squared * 2)] = - Φn / (c squared * 2) =
= (K*M) / (c squared *r * 2)

That is to say, half of the value predicted in the formula of the first book.

Back to the top