No, the distant starlight problem is not the same as the horizon problem

Posted at 10:00 on 08 January 2018

The featured image in the header of this post shows two colliding galaxies, formally known as NGC 4676, or more colloquially, as "The Mice." It comes from NASA's Astronomy Picture of the Day website.

They are about 300 million light years away, more than 100,000 light years across, and colliding with each other at somewhere in the region of 200 miles a second. The "tail" of the right hand galaxy is the result of tidal forces stretching it out across vast reaches of space as it collides with its partner. A quick back-of-the-envelope calculation tells you that that tail must have taken at least one hundred million years to get spread out like that.

Distant starlight is a massive problem for the young-earth timescale. Not only must light have taken billions of years to reach us from distant galaxies, but when it arrives, it shows clear evidence of processes that must have been going on for millions of years already. Astronomy PhD student Casper Hesp has a series of posts on the BioLogos website where he examines the evidence from distant starlight in considerable detail. Another example that he cites is relativistic galactic jets.

On a related note, the oft heard YEC claim that galactic spiral arms could not persist for billions of years is not true. Spiral arms have been well understood since the 1960s to be waves of high densities of stars within a galaxy: a theory that has been confirmed by computer simulations showing them to be extremely stable. The Wikipedia article on density wave theory has some animations showing clearly how it works.

How can we see distant starlight in a young universe?

If you read the attempts by YEC organisations such as Answers in Genesis, the Institute for Creation Research, Creation Ministries International and others to address the problem of distant starlight, you'll find that they all claim that standard Big Bang cosmology has exactly the same problem:

It’s interesting to note that big bangers have exactly the same problem. That is, the background radiation temperature is almost uniform, to one part in 100,000, at about 2.725 K, even when we look in the opposite directions of the cosmos. Since the big bang would predict hugely different temperatures, how did they become so even? Only if energy was transferred from hot parts to cold parts. However, there hasn’t been nearly enough time for this to occur even in the assumed time since the alleged big bang—see the instructive article Light-travel time: a problem for the big bang by Ph.D. astrophysicist Jason Lisle.

There are just two problems with this argument.

1. It doesn't answer the question.
2. It isn't true.

This problem, also known as the horizon problem, is indeed a real one. But it isn't even remotely similar to the distant starlight problem. The only thing that the two have in common is the problem of light travel time. Beyond that, the differences are so massive that to call them "exactly the same" is absurd.

The first, most obvious difference is scale. The horizon problem concerns distances of billions of light years: the size of the visible universe. The distant starlight problem, on the other hand, concerns distances of just six thousand: a fraction of the size of our galaxy. That is a difference of six orders of magnitude. It is the difference in size between a mountain and a molehill.

The two involve completely different eras of cosmic history, and completely different laws of physics. The horizon problem only concerns the first 0.002% of the age of the universe (300,000 years). The distant starlight problem concerns the entire history of the universe almost right up to the present day. The horizon problem operates at scales where the laws of physics are not fully understood, and that lie at the very limits of what we can explore experimentally and what we can theorise about. By contrast, the distant starlight problem concerns laws of physics that are well established, far more readily accessible to astronomers, well within the capabilities of modern measurement, and mathematically far more straightforward.

Omphalos, oh omphalos

The other big difference between the distant starlight problem and the horizon problem is that the distant starlight problem requires the creation of evidence for a history of events that never happened. The horizon problem does not.

Young-earth astronomers have made several different attempts to solve the distant starlight problem. These include Barry Setterfield's c-decay; Jason Lisle's anisotropic synchrony convention; and Russell Humphreys' white hole cosmology. All of these make predictions that are not observed in nature; some of them descend into absurdity; and none of them can account for features of the cosmos that show evidence of a lengthy history, such as galactic collisions and relativistic jets.

Now the horizon problem does have a possible solution in cosmic inflation, which proposes that in the first 10^-32 seconds after the Big Bang, the universe went through a period of dramatic expansion. To be sure, inflation is a bit of a mind-bender, and it does sound a bit whacked out, but it is a valid solution to the Einstein field equations, and many (but not all) cosmologists believe it to be the correct one.

But even if inflation turns out to be wrong, the universe can be explained in terms of the Big Bang being very finely tuned. Many scientists find that a bit of an ad hoc explanation, but there's nothing theologically objectionable about it, and in fact, it would be compelling evidence for design. But there is no false history involved, the universe remains the same age as it appears to be, and the integrity of the Creator is upheld.

The fact remains that, far from being the same as the horizon problem, the distant starlight problem is in a completely different league altogether. To claim that the two are the same, when they are separated by six orders of magnitude, is patently absurd. It simply doesn't make sense.