Skip to Content
16Nov 2012

BOSS III Dark Energy Survey: Call Wittgenstein! News for which plain English (so far) fails.

BOSS III Dark Energy Survey: Call  Wittgenstein! News for which plain English (so far) fails.

I've been puzzling for hours today, well into a Friday afternoon, over a technically arcane but important bit of progress in dark energy observational cosmology. Several news outlets wrote it up. All agree that the remarkable Sloan Digital Sky Survey, which is now in its SDSS III stage, has provided to its large team of users a terrific new tool piecing together the history of how the universe has expanded. It was about 15 years ago that news broke that the cosmos, rather than yielding to its gravity and slowing as it expands, is instead speeding up. Recent Nobel Prizes recognized how important that was. The project in news this week  aims to move the ball forward by charting the history of that reversal of cosmic acceleration's vector.

     It is called the BOSS project, for Baryonic Oscillation Spectroscopic Survey and that relies largely on the Sloan Survey's pretty little telescope at New Mexico's Apache Point Observatory. It released some early results of its spectroscopic analysis of the light of thousands of very distant quasars. The light carries the fingerprint of all the hydrogen gas it encountered in the billions of years it has spent reaching us. And somehow, the news reports tell us, from this the astronomers are starting to see how dark energy 11 billion years ago was already resisting gravity, keeping the universe expanding a little faster than one might expect. About all that had been known from earlier surveys, mostly of supernovae behavior as seen at various distances, is that around seven billion years ago the deceleration of the Big Bang's expansion's turned to acceleration - throwing glactic clusters farther apart at an ever increasing pace and into a far future of cold, infinite void. Now, it appears, something of the earlier push me - pull you contest is being inferred.

   But man oh man, nothing I've read - not press release, not news stories derived from it and the raw paper (see Grist) and not additional reporting gives a sense, in layman's terms, what these astronomers and theoretical astrophysicists have done to get their answer. Oh, they have jargon. They include of course baryonic oscillations because that's in the name of the project, and Lyman-Alpha forest because that's the spectroscopic fingerprint of intergalactic hydrogen which indicates how much hydrogen was going how fast, and where and when.

    But that's about it. This isn't like quantum mechanics or relativity or string theory where one has to do the math to really appreciate it. Separately the phenomena - early vibrations kind of like sound waves, absorption lines, etc - are describable. It's the whole package that eludes comprehensible description. Hence the allusion in this post's hed to Wittgenstein, the Austro-British logician who puzzled out examples of ideas that could not be expressed without logical contradiction or ambiguity in some systems of language while others might do the job. Science writers pride themselves on jugglng allusions, metaphors, analogies, and deft quotes at least to provide lend a sense of what a complicated science story is about and some of the scientific method employed to put it together.

   Maybe it's just me. This post points no fingers. It simply shares my struggle to internalize this news and the science behind it. But take a look at a few examples from among news stories and at the press releases and the paper and stuff like that. If you're a little fuzzy on your data deconvolution algorithms and acoustic baryonic oscillation, do YOU get it? What feature in these schmeared-together Lyman-Alpha spectroscopic features combined with baryonic oscillation hints and all coming from the same pixels simulateneously told them not only how fast gas at varying distances was going, but when? I think my brain has multivariate paralysis.


  • Symmetry Magazine - Kelen Tuttle: BOSS collaboration meausres expansion of the universe 11 billion years ago ; Symmetry is not quite general media, but is more of outreach by Fermilab and Stanfords Linear Accelerator Ctr and aimed at a very techie crowd. The piece says what they did, but not how they did it. One gets the sense that these baryonic accoustic oscillations mediated the large scale distribution of matter, but how does one know the exact distance to given pieces of it?
  • Astronomy Now/ Spaceflight Now - Stephen Clark: New technique brings ancient dark energy into focus ;
  • - John Hollenhorst: Utah dark energy study hints at the end of the universe ; I love this story, not becuase it is so good but because this TV newsman in Salt Lake didn't even try to relay what the basic importance of the news is, but charged in anyway. He interviewed the Univ. of Utah man features in that school's press release, didn't say anything at all about what the larger project is by name or other institutions, and writes it as though it's chiefly a Utah-based effort and as though debate between eternal expansion or big crunch is still running hot in the halls of science. But the story does get the gist of the issues involved and the general spookiness of contemporary cosmology.
  • BBC - Jonathan Amos: Quasars illustrate dark energy's roller coaster ride ; Amos does make a stab at how the baryonic oscillations left behind indications of distance - a "preferred scale" of galactic and supercluster distribution. Still don't get it.
  • : Distant Black Holes Map Universe's Dark Energy History ; The lede refers to "Extremely bright black holes," which is pretty much what a quasar is even though it's not the supermassive black hole itself that's bright, it's its sloppy eating habit and the blowing, glowing leftovers.
  • Guardian (UK) Stuart Clark: Dark energy map puts the squeeze on dark matter ;

Grist for the Mill: SDSS-III Press Release ; Lawrence Berkeley National Laboratory Press Release ; University of Utah Press Release ; arXiv paper preprint.

- Charlie Petit





I for one have followed the remarks. Thanks one and all for their serious tenor. That goes especially for the sdss press officer (Jotdan Raddick?) and the inside view of the press release's gestation. / Charlie Petit

I don't suppose anyone is still following this thread, but I have to return to it anyhow. If anything,  here we have a great example of how trying to "explain" things using metaphors is fraught with trouble. While it's true that a very good metaphor can in some cases help even scientists find their way through a forest (and since I'm using a metaphor here...), metaphors have a lot of power to simply mislead and confuse. People inevitably try to "reason" using metaphors, and in the effort can hardly avoid becoming totally lost due to missing details that the metaphor doesn't capture.

This is why a roller coaster metaphor is not good for "explaining" dark energy. Dark energy is still a hypothesis, though one that fits observations pretty well, as these SDSS results show. It arises very naturally from the equations of general relativity applied to the situation of an expanding universe. Of course, showing people even the simplest equations tends to be a major turn-off, and the equations in question aren't quite the simplest, though they aren't all that bad.

The problem is that dark energy, if it is in fact what the equations suggest, is something unlike anything that any person has direct experience with. That's because the quantity of dark energy in a region of space increases in direct proportion to the volume of space, while the amount of matter or ordinary energy doesn't change at all. So the density of dark energy remains constant as space expands, while the density of ordinary matter and energy decreases (in inverse proportion to the volume increase). And consequently, the strength of the effect of dark energy eventually surpasses the strength of the gravitational effect of ordinary matter.

The equations predict that there is a certain point in the universe's history. Before that point, the gravitational effects of ordinary matter were dominant over dark energy. After that point, the opposite is true. That point was about five or six billion years ago. For about the first  7 or 8 billion years after the big bang, the universe was still expanding on the momentum from the original event, but it was slowing down. It was decelerating like a roller coaster on an uphill track, so at least the metaphor describes what was happening, though the mechanism was entirely different. After the critical point, the expansion accelerated again. This too is what the metaphor suggests. However, the important point is that the equations that control the motion of a roller coaster car are entirely different from the equations controlling the motion of the universe on large scales.

The main significance of the BOSS results being reported is that the observations show the equations of general relativity that contain dark energy (of a particular type: a cosmological constant) fit the data very well - much better than the much simpler roller coaster equations (which, in any case, lack a physical basis for being applied).

Why is this important? Because when funding agencies look at this kind of research, if the research is explained well, the agencies (and Congress, etc.) can clearly see how effective this particular BOSS research has been: it clearly supports certain types of equations describing what is happening and significantly narrows down the range of possible explanations. Since research dollars are very scarce these days, it's very important to recognize types of research that are effective, in contrast to other type which, perhaps, are not so effective. In reading the press release, I see that some of the important points were made - but the imporance doesn't come across as well as it should. This is understandable, since there are many concepts that need to be explained. However, a lot of the "background" is not especially new, so it shouldn't need to be handled from scratch.

Obviously, a science writer is not going to put together a good explanation of this research on a tight deadline. Nor should the writer have to. The SDSS news office should be totally aware of how to explain the science behind the research SDSS enables - after all, it's been in progress for a number of years. There's no need or excuse to do it at the last minute. Most readers, it is true, won't care about all the exact details, especially if a few thousand words are needed to do justice to the subject. But the words need to be written down somewhere so they can be called up when required. Press release writers and journalists who report on the subject should feel obligated to read the long form explanation in order to inform what they write in their 700-word story. Any writer who aims for something longer to be published in a suitable space (Wired, NYT magazine, New Yorker, Smithsonian article, whatever) can take enough time to do it right, especially if the research insititution news office has already done its own long form write-up.

Thanks for the fascinating discussion topic! I'm the press officer for the Sloan Digital Sky Survey III, so I wrote the SDSS-III press release linked above. We struggled with these issues while writing the release, and I definitely felt like Wittgenstein while writing it (although I would have never thought of Wittgenstein as an analogy for that feeling!). We did the best we could, but I definitely thought the release was a bit overcomplicated.

The problem, as you point out, is that while each step in the logic of the discovery is not string-theory complicated, there are so many steps that it's hard to tie them all together in a story. Furthermore, the steps are more or less unrelated - the technique of the Lyman-alpha forest is a completely different explanation from the implication of the accelerating vs. decelerating universe.

Early in the process, we came up with the idea of a roller coaster going up a hill (decelerating) and then down (accelerating). It was a fun analogy, but it never really took hold throughout the release.

Kudos to that TV reporter in Utah for giving it a go. This must have been a tough story to present as breaking news. I’m fortunate I once enjoyed the luxury of writing for a weekly back in the olden days when you didn’t have to plop something on a website the instant someone announced a new finding.  

BOSS is an interesting project and there’s room for a good feature story putting it all in context. It’s easier to be clear when you have a little time to work with.

Oh yes, I have given up on natural language, let alone pop journalistic language, to convey anything with information value.  It is wholly weighted down with local social norms and ideology.  In brain research the words: reward, choice, decision-making and even emotion, personality, etc have almost become meaningless in discussing brain and behavior.

It appears to be a law of nature that "Information is expensive." Media, pop culture, journalism, by definition, are paid time, money and attention to be inexpensive.  Hence, of little, if any, information value.

The kumbaya notions of science journalism need some heavy rethinking -- and evidence.  What a novel idea!

I am a marketer, pro communicator and brain research geek.  My expereience is thre is no such thing as pop science.  To understand you have to study and know some basics.  That is counter to most pop ideology but is our experience and we have tried.

On the other hand Steve Pinker is trying harder than anyone to move the mountain.  See -

We now restrict our work to semi-pro and pro readers.  Even then it's tedious and, mainly, argumentative work. 

The human brain has no facility with complex topics.


The key thing here is determining not how fast gas was moving at particular times (i. e. distances from Earth), but the density of the gas at different distances.

Here's how I would explain it:

Baryons are the type of particle that make up most of ordinary matter, as found in stars and planets. An acoustic oscillation is a small periodic fluctuation in the density of matter - in other words, a sound wave. Matter in the very early universe vibrated like a ringing bell, at a characteristic frequency, and this initiated density fluctuations in baryonic matter (mostly hydrogen) that have propagated through the universe ever since. The wavelength of these fluctuations can be predicted from conditions in the early universe, but it has steadily increased over time, along with the general expansion of the universe. For most of the universe's history, the wavelength has been very large - on the scale of millions of light years. Though the wavelength has been steadily increasing, the rate of increase has varied over the history of the universe, due to the varying effect of dark energy.

Since the wavelength of matter oscillations is simply the distance between successive density peaks, mapping matter density peaks reveals what the wavelenth has been at any stage of the universe's expansion. The average density of matter at particular distances from Earth can be inferred from the effect it has on light from quasars that are more distant than the location where the density is to be measured. Light that was emitted from quasars in the ultraviolet range carries a distinctive series of emission lines known as the Lyman-alpha forest, so named because of their origins in electron energy level transitions in atoms of very hot hydrogen gas. The degree to which the Lyman-alpha emission lines are absorbed is a direct indicator of the density of intervening hydrogen gas. By comparing the amount of absorbtion of quasar light from quasars in the same direction but different distances, the density of intervening hydrogen can be estimated - which enables peaks in the density distribution to be located.


Login or register to post comments