Skeptoid PodcastSkeptoid on Facebook   Skeptoid on Twitter   Skeptoid on Spotify   iTunes   Google Play

Members Portal

Support Us Store


Get a Free Book



Crashing galaxies, or crashing research ...

by Bruno Van de Casteele

September 8, 2013

Share Tweet Reddit

Donate Sometimes I file away interesting articles to catch up on later (or hoping that I can blog about it). For instance, this articlefrom the beginning of Julyabout research that proves that Andromeda and our Milky Way collided with each other 10 billion years ago. The linked article puts a question mark behind it, but the same article appeared in other forms with a headline that makes it seem like a fact (for instance here). The original article can be found on arXiv. But this article caught my attention when I read it in more detail, and not in a good way.

The problem is not with the titles. The problem is actually about the research and how the researchers presented it in the press. First, you have to understand that these results were obtained by using MOND, Modified Newton Dynamics. This is an alternative theory that wishes to explain the rotational observations of big galaxies by adapting Newtons theory, instead of dark matter in our current theories.When you use MOND, you can also apply it to other large scale phenomena, like how galaxies interact with each other. That is exactly what the team of Hongshen Zhao from the University of St Andrews set out to do.

Let me first state, for the record, that I think MOND is not a pseudoscience. I applaud the fact that there are competing theories to the more established ones, and that, in this case, the hypothesis is not just applied to account for the anomalies, but wishes to provide more "positive" proof for the validity of it. Those efforts should be lauded, cherished and funded.

But there is something tricky here. As the articles report, it almost seems certain that these two galaxies collided in the past if you "just" accept the other theory. However, MONDwas initially meant to counter for the "invisible" dark matter. We found more and more observational proof of this, and it very elegantly explains not only the rotation of one galaxy, but also how galaxies as a whole behave. According to our current understanding, such a collision could not have happened so long ago without us eventually merging with Andromeda.But that fits very well current observations!

Furthermore, MOND is now targetting the existence and distribution of dwarf galaxies. A member of the team, Pavel Kroupa from Bonn University, sees no "other explanation", but that seems at best a bit lazy. There is valid research going in (more than 3000 articles on "distribution of dwarf galaxies" in Google Scholarthis year), and it certainly isn't the achilles' heel of the established paradigm. Furthermore, as I read the conclusion of the original article, it seems that the observed dwarf galaxies are just a "possible" result of this colliding into each other, and not proven. It shouldn't even be part of the conclusion, in my opinion.

So while the research sure has merit (if only to challenge the gaps in our current understanding), the reporting and the explanations given by one of the team members are over the top. There is talk of a "smoking gun" and "rewriting the history of the cosmos", but that is really overblown given the partial model used (two-dimensional, no other galaxies or clusters). I can only speculate that the triomfant language is maybe meant to secure funding to calculate the creation of these dwarf galaxies in a more comprehensive model. However, by reporting on their current findings like this, I think they are doing themselves and MOND a disservice. If they are right (and why not?), the hypothesis will eventually win out.

Oh and if you like colliding galaxies, don't forget that we will eventually collide with Andromeda within four billion years. But don't worry... we will pass mostly through one another without violent crashes, so no need for helmets.


by Bruno Van de Casteele

Share Tweet Reddit

@Skeptoid Media, a 501(c)(3) nonprofit








Want more great stuff like this?

Let us email you a link to each week's new episode. Cancel at any time: