Star-Planet Interactions, and Jupiter Analogs

Waaaaay back in 2015 the International Astronomical Union held its General Assembly in Honolulu. I went and gave a review talk on star-planet interactions at a Focus Meeting.

One nice thing (in the long run) about these Focus Meetings is that they generate proceedings that get published. It’s sort of old-fashioned now, but it’s still nice to see these proceedings because they often contain things not in refereed papers: preliminary, unrefereed results that turn out to be important later, and overarching but concise syntheses of lots of work in a way that is useful for understanding but not really appropriate for a refereed article on novel research.

(I write “in the long run” above because having to actually write the proceedings can be a pain, and because they seem to take fooorrrreeevvvveeeerrrr to finally get published.)

Brendan Miller

Well, I was going through my CV for my end-of-sabbatical report (7 more days!) when I remembered that Brendan Miller and I put in a proceedings for the 2015 summer meeting!  Whatever happened to it?  Turns out it was published a while ago and somehow I missed it (which is weird because I have a copy of that book on my shelf…)

Anyway, our contribution is now belatedly on the arXiv.  Here’s what’s in it:

We really want to study the magnetic fields of exoplanets. It seems sort of hopeless—magnetic fields don’t have that much energy and it’s hard enough to figure out a planet’s mass, much less this little detail—but there is hope.

One hope is that close-in exoplanets will have their magnetic fields interact with their host star’s magnetic fields, causing magnetic activity on the star that we can detect in the calcium H&K lines. There had been suggestions in the literature that this was happening, as magnetic “hot-spots” beneath close-in planets rotated in and out of view, but follow up of those systems found the effect to be difficult to reproduce.  I think it was noise.

Another hope was that there was an overall increase in the level of activity in stars with close-in exoplanets.  If you took a sample of stars with and without close-in planets, were the ones with close-in planets more active?  Turns out that’s hard, because there are lots of biases in the way we detect close-in planets (via transit) that might make it more or less likely to find them around active stars in the first place.  Brendan and I wrote a paper where we looked at the evidence (and gathered some ourselves) and concluded there’s no signal to we can dig out of all of the noise.

But there are clear cases where there is star-planet interaction, just by another route: close-in, very massive planets seem to be able to spin up their stars, which makes them more magnetically active.  That probably drives the small amount of correlation we do see.

Then Brendan took a look at WASP-18, which should have one of the strongest planet-induced activity levels around if that’s a thing, and found it’s not elevated in X rays.  Bust there, too.

One thing we did not have time or space to touch on in the article was the one way that magnetic fields do seem to have been detected, via bow shocks., which is a shame but was fortunately covered later in the session.

There is one more bit in the paper that has been dribbling out slowly over the past few years, too. One of my earliest interesting papers was announcing the discovery of the first really good Jupiter analog HD 154345 b.  It’s around a G star, has about an 8 year circular orbit, and is around one Jupiter mass.

One gotcha is that the planet has the same orbital period (and phase!) as the star’s magnetic activity cycle. That’s not too surprising: stars’ cycles tend to be around 10 years, and so some will inevitably have planets at similar periods. The phase matchup is a further inevitable coincidence. After all, our stablest stars, like σ Draconis, have big strong magnetic activity cycles and those don’t create phantom planets in our radial velocity measurements.

Or so we argued in the paper. Well, since then, the coincidence between activity and RV has been getting better and better, and as early as seven years ago I had been conceding that this might be a rare, strong activity-RV coincidence.  I mentioned it in at the first EPRV Workshop (you can see it in the slides here) and again at the 20th anniversary of 51 Peg conference in Haute Provence.

Well, here it is again, in our review:

This is one of those cases where I really should get this into a refereed paper, but I’m busy, and more data will make the case stronger, and retractions are hard to get motivated to write.  Anyway, this has been out there for a while in unrefereed form (and actually disputed! though I still think the planet is probably wrong) but I hope to get it properly written up this fall.

Anyway, that’s the news from Lake Wobegon, where all the planets are Earth-like, all the objects are Rosetta Stones, and all the signals are significant.

 

Clickbait and Sensationalism

Journalism has been in trouble since… well maybe since forever but it certainly feels more precarious since the World Wide Web destroyed most print journalism as a viable stand-alone business. The New York Times and Washington Post have hung on with quality journalism, but not without moving to a heavy online presence and worrying a lot about generating “clicks” (at least on the business side).

One way to survive is to generate lots of those clicks, and that means “clickbait”—provocative headlines that dare you to ignore them. Whether you find this to be outrageous or just a fact of business probably depends on how exposed you have been to it.  I imagine most journalists take it as a given that headlines and ledes must generate clicks and scrolls in order for them to have jobs, but that what really matters is the meat of the article.

It would be nice if it weren’t so, especially when it’s your science being hyped. Yesterday I wrote about Sabine Hossenfelder’s Forbes article and along the way defended it against accusations that it was “sloppy journalism” and representative “the frenzied sensation-driven nature of mainstream publishing” mostly by linking to it, quoting it, and being incredulous at those descriptions. Some have disagreed with that assessment:

First of all let me sympathize: it is mortifying to see your research in print in a way that sensationalizes it.

Next, let me point out that this is not even close to yellow journalism. I know yellow journalism, and Forbes, senator, is no yellow journalism.1

So maybe it’s because I’ve seen a lot worse and I’ve written for popular outlets and my student Kimberly Cartier wrote a thesis about science communication that I’ve become inured to the way that headlines beg for clicks, even if they don’t represent the meat of the article well.

But I think in this case the Forbes article is actually not in that camp. With the benefit of hindsight or insider knowledge, it might look overblown, but consider:

  1. At the time it was written, there was no LIGO response to go on.
  2. Hossenfelder contacted LIGO and they had no formal response—and said they planned none!
  3. Hossenfelder then got a third party to comment, and they found “the results “quite disturbing” and hope[d] that the collaboration will take the criticism of the Danes to heart. ‘Until LIGO will provide clear scientific (!) explanation why these findings are wrong, I would say the result of the paper to some extent invalidates the reliability of the LIGO discovery.’”
  4. We all know that the scientific literature is often understated, especially when suggesting someone is wrong. An abstract that concludes “A clear distinction between signal and noise therefore remains to be established in order to determine the contribution of gravitational waves to the detected signals.” has a pretty clear meaning: the GW detections are in jeopardy.
  5. After all, the strongest signal was 5.1-sigma.  If there is any reason to think the noise is underestimated then the signal drops below the 5-sigma level.  Now I appreciate that the difference between “significant” and “not significant” is not Itself statistically significant,” but the “New York Times threshold” is actually the one place where which side of 5-sigma you are on absolutely matters!

Given all these items, I think the headline and fInal introductory sentence of the Forbes article is fine:

Was It All Just Noise? Independent Analysis Casts Doubt On LIGO’s Detections

…But what if there wasn’t a signal at all, but rather patterns and correlations in the noise that fooled us into believing we were seeing something that wasn’t real? A group of Danish researchers just submitted a paper arguing that the celebration might have been premature.

Especially given the 5.1 sigma issue, I think that’s totally fair. The paper did cast doubt, none of Hossenfelder’s followup investigations dispelled that doubt, and “celebration might have been premature” is an accurate description of the implications if the detection was actually only, say 2.5-sigma (and Creswell et al. implied it could be much lower than that, IMO).

Anyway, YMMV on this. We can agree that it would be nice if popular journalism were as sober as The New Yorker and as popular as cat memes, but I think we can also agree that that will not happen.

The only practical solutions I see to the larger problem are:

  • Get scientists better at science communication.
  • Appreciate those that are good at it and see them and the work they do as an important part of our profession (the aforementioned Kimberly Cartier included a chapter on this as part of her PhD thesis in Astronomy & Astrophysics—I hope in the future this will not seem novel).
  • Make sure that when our work pops up in the media we know how to manage it. Especially a large, high-profile project like LIGO should have1 team members (or access to professionals) that know how to quickly manage stories.

After all, if LIGO had told Hossenfelder that it was planning a response that completely addressed the Creswell et al. paper and showed it was based entirely on poor analysis, her article would have had a different conclusion and different tone (if it had been written at all!). They had an opportunity to get in front of it and shape the Forbes article, but they did not.

But it all turned out OK: their eventual (though unofficial) response is solid and succeeded because they put it on Sean Carrol’s popular and respected blog instead of waiting for the peer review process. Now the story can be “nothing to see here, move along” and, as a bonus, the referee of the Creswell et al. paper has a great template for their review. Double bonus: now the world understands LIGO better!

1For the kids: Bentsen-Quayle.

[Edit: Hannam responds (click to expand):

]

1I don’t mean to imply LIGO doesn’t have such people—on the contrary I mean that the (by all accounts very good) people it has are essential.

Outreach and Response

What good are blogs?

The LIGO collaboration recently made headlines, made an epoch, and earned itself an all-but-certain Nobel Prize, for the detection of gravitational waves.

Recently, an unrefereed paper appeared on the arXiv questioning the strength of the detections. In brief, LIGO uses two detectors separated by 2,000 miles to filter out noise by looking for coincident signals: real gravitational waves will affect both sites equally, but local sources of vibration should be uncorrelated between the sites.  The Danish authors, who include a scientist name Andrew Jackson, took some public LIGO data of the detection, did some analysis of it, and found that the noise appeared to be correlated between the two sites.  At the very least, they argue, this means that the LIGO collaboration has overestimated the strength of their signal.

It’s a provocative claim: that a major milestone in physics could be a mistake, revealed by a relatively straightforward analysis that any physicist could understand. When I heard about it, I thought “that’s probably wrong, but I’m curious why it is that the sites have correlated noise and how the LIGO team deals with that.”

Sabine Hossenfelder is a theoretical physicist, blogger, and freelance science journalist. She apparently had similar thoughts to mine, and used them as the basis for a column in Forbes on the topic. It’s a nice piece of science popularization, that explains the issue in an accessible way.  She, very responsibly, went to the LIGO team for a response:

Jackson is no unknown to the LIGO collaboration. Upon my inquiry with a member of the LIGO collaboration what to make of the paper, I got the annoyed reply that the collaboration’s management recommends to “respectfully respond that we have talked at some length with the group in the past and do not agree on the methods being used and thus with the conclusions.” Another let me know that a response is not planned.

She also walks the reader through some of the problems with the claim.  I found the column illuminating, but wished I had a better explanation of the issue. I was intrigued!  Hossenfelder concludes:

Making sense of somebody else’s data is tricky, as I can confirm from my own experience. Therefore, I think it is likely the Danish group made a mistake. Nevertheless, I would like to see a clear-cut explanation and “they did something wrong” is too vague for my comfort. This is a Nobel-worthy discovery and much is at stake. Even the smallest doubt that something is at odds should be erased.

Right! But even beyond the issue of how if they want their Nobel they should make sure there is no doubt, there’s also a general outreach angle: gravitational waves are a very popular topic, and this paper, however annoying, had the effect of raising interest in a particular aspect of the science. This paper provided the LIGO collaboration with an opportunity to cash in on that level of interest and explain this particular aspect of the science.

And they did. A member of the collaboration, with the blessing of the collaboration, wrote a guest post on Sean Carroll’s popular blog.  It was just what I hoped to read: an accessible (to me, anyway) discussion of why the Danish group’s analysis is almost certainly wrong and naive, along with a quick tutorial on how LIGO makes sure it doesn’t make similar mistakes.

The whole thing to me was a good example of how various levels of science communication can work: the arXiv (for better or for worse) provided a formal forum for a team to make a scientific claim of high visibility before peer review; the science column provided a way for a professional scientist to engage the public in the issue; the science blog provided a way for the team to make an informal but quick and almost definitive response to what was apparently a straightforward mistake by the Danish group (thus illustrating why one shouldn’t make provocative, unrefereed claims on the arXiv: you’ll usually end up being very publicly wrong).

Then there was a blog post by Mark Hannam on the whole episode.  Now, I get that the LIGO team is understandably frustrated by this sort of high profile sniping by a team that apparently didn’t know what they are doing, and annoyed that they have to spend time putting out these PR fires. The mature response is to turn it into a teaching moment, and they did with the Carroll guest blog post. Now the broader community understands better, the Danish team’s mistake is laid bare, and everyone knows more physics than they did before.

But Hannam doesn’t reserve his fire for the Danish group.  He actually says the thing that annoys him most about the whole thing is Hossenfelder writing it up in Forbes! He refers to her “sloppy journalism” and “the frenzied sensation-driven nature of mainstream publishing”. Did he read the same article the rest of us did?  Take a look for yourself.

Hannam is annoyed that the “controversy” played out it real time and not at the pace of peer reviewed paper. I expressed my eye-rolling at Hannam’s post on Twitter, and got some pushback, mostly because in the confines of 140 characters it looked like I was attacking the LIGO collaboration. It went over to Facebook, and a lot of people disagreed with me.  Some excerpts of rebuttals:

“the methods used for the discovery have been laid out in excruciating detail”

Well, yes, but saying that interested people could figure out the problem by reading your papers is terrible popular science communication; it’s much faster and more efficient for someone to take some time to explain it to everyone briefly than to expect everyone else to take a couple of days to digest the papers and figure out for themselves what the Danish team did wrong.

Of course, you don’t have to communicate your science well in general, but it’s sort of an obligation when your project is on the front page of the New York Times and in the running for a Nobel.

“Younger scientists may think it is cool to have open discussion with non-peer-reviewed arxiv-postings but that’s busy work that takes away from more important analysis and when I was a student/postdoc that’s what conferences were for”

The quick answer is that blogs are cheaper, more far-reaching, and faster than conferences, so why prefer conferences for this particular item?  And it is only “busy work that takes away from more important analysis” if you think communicating your science is not important. If you think it’s part of your job, then you know you have to do it anyway, so it makes sense to focus your efforts on items that already have the public’s attention, like the Danish team’s paper.

Ok, let’s consider an analogy. Recently, we did a press release for a paper that got picked up by a few people, and there were several articles, some with open online comments. And some of those comments (trying to think of a non-“scornful” way to say this) strongly disagreed with our work while exhibiting a lack of familiarity with the subject. Would Jason Wright and James Guillochon suggest that I “missed a teaching opportunity” by not engaging with the comments?

Well, yes, obviously, but not a very big one. Responding to every comment on every article is obviously inefficient and doesn’t scale. But when the whole world is watching your team and your results are influencing in billion-dollar space missions decisions and a Nobel Prize hangs in the balance, it doesn’t seem unreasonable to have some kind of public response to news stories on an accusation from your colleagues that you’ve made a huge mistake.

This is actually a great argument for science blogs: it’s a way to quickly, very publicly make an arbitrarily-detailed response to things like an unrefereed arXiv post. They let you balance the time you put into the response to the claim you’re responding to.

For instance, instead of putting out fires as they come up on Twitter and Facebook as a result of my off-the-cuff tweet, I can have a blanket response to all of them in one big post I can link to.  It’s the same reason that people hold press conferences instead of answering phone calls from reporters all day (but on a much smaller scale, of course).

As I said, if I were the LIGO team I would be annoyed by the episode, but the things that would annoy me the least are that the world showed an intense interest in my work, that I had to explain my science to that interested audience, and that I got to show up a gadfly on a big stage.

[Edit: More responses and details in my next post]

[Edit: Hannam responds (click to expand):

]

 

Tabby doing a Q&A on the WTF star on Twitter

Tabby just did a 20-questions-and-answers thing on Twitter.  I found it hard to read the whole thread, so I’ve compiled it here.  Enjoy!

 

 

Two New Tabby’s Star Papers

Amidst the huge task of collating all of the data coming in from the May 20, 2017 dip, two papers have hit the arXiv.  I don’t have any updates on the data from the dip (we haven’t had time to do any detailed analyses yet), but the live chat I did on Friday is still mostly valid:

except to say that the dip has maybe ended:

Today there are two new papers on the arXiv on the subject.  I haven’t had time to do deep dives on them (and neither is refereed yet) but here are my hot takes:

The first is by Ballesteros et al. (MNRAS, submitted) and they try to model the dips with a gigantic planet with a huge ring system and huge swarms of trojan asteroids.  In other words, their model puts a lot of stuff in a 6 au orbit around the star, which is far enough away that it would be pretty cold.  They point out that the deep, asymmetric dip at Kepler day 793 occurs about half way in the middle of a pretty quiescent period for Tabby’s Star.  They associate the other dips with swarms of trojan asteroids—asteroids in the same orbit as the planet but leading or trailing the planet by 60 degrees.

Some strengths of the model:

  • They claim that they can model the deep D793 event as a giant (0.3 solar radii!) planet with a tilted ring system and that they will do this in a later paper
  • They get the overall pattern of the dips explained: Kepler just caught the back of the pack of leading trojan asteroids when it started observing, then the planet at day 793, then the trailing swarm at the end of its mission
  • In what must have been a hastily written addition, they attribute the May 20 event to a secondary eclipse of the planet behind the star. This comes with a prediction: the event will be no longer than the D793 event (which was actually very long), but they say no more than 2-4 days.  They say that the secondary eclipse depth could be as deep as 3% (about what we see).  I note it should also be pretty achromatic, unless the reflectivity of the planet is a strong function of wavelength.
  • They emphasize that their model appeals only to likely, conventional astrophysics (though when it comes to 0.3 solar radius planets and a Jupiter-mass of asteroids in a swarm, your mileage may vary on that one).
  • They have a really nice diagram!

Some drawbacks:

  • They need a lot of asteroids: they don’t actually say how much, but the number they do give is huge: over a Jupiter mass of them!  It’s not clear to me how stable such a swarm could be co-orbital to an actual planet.  Part of the reason Jupiter’s trojan asteroids work as they do is that they don’t really perturb Jupiter. Also, how do you keep a Jupiter mass of material from collapsing or falling into the planet?  Also, where would you get a Jupiter mass of rock?!
  • They cannot explain the secular dimming seen by Montet & Simon and Schaefer, which they say must have a different cause.
  • They do not confront the infrared and mm upper limits, especially those of Thompson et al. (whom they do not even cite) that put no more than a millionth of an Earth mass of dust hotter than 160K.  I would think that an asteroid swarm dense enough to have an optical depth near 1 along some lines of sight (22% dips!) would also generate some serious dust, as would those rings.
  • They will need a pretty strange sort of planet to have a detectable secondary eclipse out at 6AU.  They claim that a Bond albedo of 0.34 will do it, but my back-of-the-envelope calculation says no way this could work (a perfectly reflective 15 solar radius circle (for the ginormous rings of this planet) at 6 au intercepts about 1 ten thousandth of the stellar flux, not 3% of it).  If it’s really emitted light then it should be pretty red, so the May 20, 2017 dip should be hard to see in the blue.
  • I think the slopes of the dips are too steep; material at 6 au moves pretty slowly. They could easily calculate this.

But kudos to them for putting an idea out there with concrete predictions!

The second paper is by J. Katz.  I’m glad to see this one in principle—Steinn and I suggested an object in the outer solar system could be responsible and hoped someone would work that out, and here’s a paper working it out!  Weirdly, Katz cites us but don’t mention our suggestion.  Anyway glad to see it.

This is a strange paper, though. There is no comment that it has been submitted to any journal to be refereed—it’s possible this is all we get.  It’s called “Tabetha’s Rings”—I don’t think I’ve ever seen just a modern astronomer’s first name in a paper title before.  Katz refers to the star as “Tabetha’s Star” which is also strange (because the star needed another name, right?).

Katz suggests that a ringed object in the outer solar system could be responsible for the dips…and not much else.  Some of the implications are worked out, but some of the math seems wrong to me (he predicts that the dips will be visible every 365.25 days from earth, which ignores the orbital motion of that object).  I kept expecting Katz to bring up the rings of asteroids but it never came up.

Anyway, I hope Katz develops this model further and describes things like the spectral and photometric properties of the dips his model implies, and discusses, for instance, the mass of the object hosting the rings (at least!). I’d really like to see a fleshed out version of this paper in the refereed literature.

OK, the kids are off to school so time to get back to the disaster area that is my inbox…

Activity from calcium

The atmosphere of the Sun (and other stars) contains calcium. It contains most of the elements, actually, just like the Earth does. As light that emerges from the sun passes through this cooler atmospher, two specific colors of very blue light, corresponding to specific transitions of electrons in a calcium ion, have a hard time getting through because they get absorbed by the calcium. These colors are “missing” from the solar spectrum, and Fraunhofer, who established much of our notation for spectral features, labeled them “H.” Later astronomers gave the two wavelengths separate names, and today we call them the H and K lines.1

The sun is “dark” at these wavelengths (this light doesn’t get through the lower atmosphere), so the much hotter upper atmosphere of the sun (the chromosphere) stands in good contrast against it, especially because the chromosphere is bright at these wavelengths (this is not a coincidence—the same transitions that make calcium in the lower atmosphere a good absorber make the upper atmosphere an efficient emitter at the same wavelengths.)

National Solar Observatory image of the sun in the wavelength of the ionized calcium K line.

You cannot see the usual “surface” of the sun at these wavelengths; that light has all been absorbed by calcium ions. In this image you are looking at the upper atmosphere of the sun.  Here, the brighter regions are hotter, and they tend to cluster around sunspots.  This is because sunspots are caused by intense magnetic fields on the sun, and these fields reconnect and deposit energy high in the sun’s atmosphere, heating it and making it shine at this wavelength. The sun has an 11-year activity cycle, and if one makes measurements like in this image, one can clearly see this cycle as the total number of sunspots rises and falls over the course of a decade.

Now, in other stars we cannot see sunspots, but we can measure the amount of H and K line emission.  Imagine this image of the sun was taken from so far away, you could not make out the sun’s disk.  The five bright “active” regions (near the sunspots) would add up to make the point of light that is the sun look brighter at this color than it would if those regions weren’t there.  This means that you could tell how much magnetic activity—sunspots and related things—was going on on the sun by how bright it was at this color. Watch long enough, and you could tell that the sun had activity cycles!

This is the philosophy behind the pioneering Mount Wilson H & K project, undertaken by Art Vaughn, George Preston, Sallie Baliunas and many others from 1966-2002.  They measured the brightness of around 100 sun-like stars for decades to watch the rise and fall of their activity levels.  The technique is now used at many observatories.

One of the big things people look for is an analog to the solar Maunder Minimum, a period from just after the discovery of sunspots by Galileo lasting about 70 years during which there were almost no sunspots. No one knows why the sun apparently stopped its magnetic cycle for so long, but if we could catch another star doing it, then maybe we could figure it out. The Mount Wilson project identified several sun-like stars with no sunspot cycles—victory!

But in 2005 I published a paper as a graduate student showing that this was actually a mistake. All of these “Maunder-minimum-like” stars had had their distances measured since the Mount Wilson project made their discovery, and most or all of them all turned out to be much farther away than expected—which means they were much brighter than we thought.  Why? Because they’re not really much like the sun—they are subgiants, not ordinary main sequence stars, and we don’t expect subgiants to have strong magnetic fields.2  So it turned out the Maunder minimum was still sort of a mystery.

But wait!  In star clusters one knows the distance to all of the stars, so one won’t get fooled by subgiants.  Mark Giampapa and others have looked at truly sun-like stars in M67, an open cluster of stars a lot like the sun, and found that some of them have calcium H & K emission way below what the sun has even at solar minimum—there they are! Maunder-minimum-like stars!

Jason Curtis, now an NSF postdoctoral fellow a Columbia University

In an amusing symmetry, my former graduate student, Jason Curtis, has looked into this and discovered that because M67 is so far away, you have to worry about another source of absorption: the interstellar medium.  This gas between the stars is very sparse—the Mount Wilson stars are all too close to have their light affected by it.  But M67 is very far away, and there is a lot of this gas in the way. This gas is made of the same stuff everything else is—including calcium!

Maybe you can see where this is going.  The calcium in the interstellar medium absorbs calcium H & K light, making the stars appear dimmer at those wavelengths, and so our magnetic activity measurements end up giving erroneously low values. Once you correct for that absorption, it turns out that there aren’t really any anomalously inactive stars in M67.

So Jason’s new paper on this topic points out that, once again, stars that we thought were good Maunder minimum stars are, in fact, not—in this case, they’re just behind more interstellar calcium than we’re used to seeing in front of nearby stars.

You can read his (single author!) paper here on the arXiv now that it has been accepted to the Astronomical Journal.


1Jay Pasachoff pointed me to this history of notation for the H & K lines. Fraunhofer did not discover them, and the “K” line terminology came much later.  Jason Curtis points me to this amusing mistake, where the letters are misinterpreted as standing for “hydrogen” and “potassium”.


2More on this, including an amusing anecdote about a “Marshall McLuhan moment” at my first colloquium, here.

Who Should Be an Author on a Paper? V: Some Errata

It looks like my post was based on the old AAS Ethics Statement, not the more recent Code of Ethics.  That’s fine, but it means the language I quoted was not the latest.  The language on who should be an author is the same, so the heart of my posts are unchanged.

But now, the Code says:

As stated in the National Academy of Science document On Being a Scientist, “The list of authors establishes accountability as well as credit,” and “an author who is willing to take credit for a paper must also bear responsibility for its errors or explain why he or she had no professional responsibility for the material in question.”

So this directly addresses one of the most common objections I’m getting (which is not really an objection to my proposal per se, as I’ve said).  Right there, in black and white, it says that authors may: “explain why he or she had no professional responsibility for the material in question.”

So this part of my proposal really isn’t very radial at all; it’s right there in the new Code of Ethics!

Also present is this new bit:

Data provided by others must be cited appropriately, even if obtained from a public database.

Which I think everyone agrees on.  My entire premise was “what if there is no appropriate citation?” and I’m asking “what does appropriate mean?” I argue that if there is nothing to cite that “counts” today, then this clause can’t be followed, so it no longer overrides the earlier co-authorship requirement.

Finally, on the obligations of co-authors it says:

Every coauthor has an obligation to review a manuscript before its submission, and every coauthor should have the opportunity to do so.

Which is a stronger statement than was in the old policy, but doesn’t affect my argument at all.

The other strain of reaction I’ve gotten is suggestions for reforming our citation and credit system, including adding levels of contributions to papers below “authorship.”  I’m all for that; my proposal had to do with what to do with the system we have in the meantime.

 

Who Should Be an Author on a Paper? IV: Practical Ethics of Authorship

Part I is here.  You’ll need to read it and prior entries for context.

Let me start this final(?) part with a formal statement of my suggestion:

In general, researchers writing a paper that uses unpublished or otherwise unciteable data they did not produce should invite the proposers/observers/producers of that data to be co-authors.

Now, there are many situations where following my co-authorship suggestion isn’t practical. Maybe there are not well defined “proposers”. Maybe the data are 30 years old and widely used. Maybe there is a timeliness or competitive issue that precludes letting the proposer know what you’re working on. Maybe the proposing team didn’t actually do a lot of work to make the observations happen. Maybe the proposer is a social pariah or one of your more important co-authors refuses to be on a paper with them. Maybe you’re on a deadline and simply don’t have time. Maybe you’re in a collaboration whose authorship rules preclude adding these people to the paper. Depending on the specifics of a situation, those might be part of completely legitimate reasons to go ahead and publish without them.

Ethics is often a case-by-case subject; broadly written rules can become outdated, or fail to anticipate pathological cases, or obviously fail in corner cases, or just be too vague to apply to edge cases.  Personal ethics also come into play: we do not all share the same values, and do not all take the same approach to collaboration. Ethics also depend on expectations of the community, and those can change.

But I think our community’s expectation and standard that we never need to include the people who took otherwise unciteable data as co-authors is wrong and should change. 

I encourage my colleagues to consider adopting a presumption that the observers/proposers of public but unpublished data should be invited as co-authors, and even taken on as collaborators early in the project. If there are good reasons not to do so, that’s fine, but those reasons should be articulated and considered and weighed against the good reasons to the contrary before a decision is made.

So before rejecting this presumption, astronomers should ask themselves:

  • Why not include them?
  • What does it really cost me to include them?
  • Why not gain a collaborator?  Why not have a longer author list?
  • What would I want them to do if the roles were reversed?

In many cases, the answers to these questions might lead authors to conclude that the producers of the data should not be co-authors, and that’s fine.

But let’s ask these questions more often.


Finally, because Josh Peek got me off on this tangent on Twitter, inspired my particular example, and is working on the MAST data policy which will guide this sort of thing, let me suggest a concrete policy for MAST, consistent with my proposal and the AAS Ethics Policy:

  1. Propriety only concerns who can see and use data. It is silent on the issues of authorship. Public data are in the public domain and anyone may download them and use them as they see fit.
  2. STScI will provide guidance to users of its data products on how to properly credit STScI and its employees for their work. This is probably something like: include the boilerplate acknowledgement, and cite such and such papers describing the instrument and analysis methods.
  3. STScI should have an internal policy for how its many scientists accrue credit (citations and authorship) for their work on projects that produce data, especially for papers produced with public data they enabled. This policy should be consistent with community norms and (hopefully) the AAS Ethics Policy (which may need to change).

That’s it!  If authors want to scoop others and not give them co-authorship, that’s not MAST’s problem (indeed, it is part of MAST’s charter to enable such scooping!).  The AAS Committee on Ethics may be interested in that author list, of course, but I see no reason (or mechanism!) for MAST to be telling its users what they can do with public domain data except publish publish publish.

OK, that’s it.  Flame on!  I will probably update this thread with more entries as good ideas roll in.

[Edit: One more post!  I linked to the old Code of Ethics.  The new one actually further supports my position, I think.]

Who Should Be an Author on a Paper? III: A Proposal

In Part I I suggested a modest apparently radical proposal. In Part II I laid the groundwork for defending it. Now, let the games begin.

To recap my concrete example, Joe and his team took public data from the HST archive as soon as they landed (this is public DDT time) and have written a paper with it.  The proposing team includes PI Candice and departed members Amber and Brie, and Candice has also written (but not submitted) a paper on the data. Should Candice offer Amber and Brie authorship on her paper (yes, I think we all agree). Should Joe offer the proposing team members Amber, and Brie authorship on his paper?

I say “yes,” because they contributed to Joe’s paper just as much as to Candice’s! The whole proposing team should be offered co-authorship. This is not current practice.

The easiest way to defend my proposal is by responding to some objections I saw when I proposed this on Twitter. I won’t link to individual tweets because I’ve rephrased some of these to be easier to rebut (hey, it’s my blog!)

But the data are public!  That means I can use the data however I want and I don’t have to include the proposers.
Also: That’s what proprietary periods are for! Once it’s over I no longer owe the proposers co-authorship.

No, data propriety only has to do with who is allowed to look at and use the data. Once the data are public, anyone can look at the data, work on the data, and publish the data. 

But that does not absolve them from their duty to properly acknowledge and credit the producers of the data. This is obvious when the data are already published. Of course you cite the origin of data you use in a paper. So ask yourself: why does the lack of a paper to cite make the procurers of the data any less responsible for their production, or you any less responsible for acknowledging that contribution in a way they get credit for?

But if they never publish their data, that’s effectively an infinite proprietary period.

Again, no: you can use and publish the data. That’s a completely separate issue from whether you have to give credit where it is due.

Why should I give co-authorship to someone that didn’t work on the paper?

Because they effectively did work on the paper as soon as you used their data in it. Since you are using their work you have to give them credit they can use.

But I list the PI’s name and the proposal number in the acknowledgements. That’s credit!

It is credit in a literal sense, but not in any sense relevant to the ethical issue here. ADS will not track it, it won’t appear on their CV or h-index, etc. It would be nice if we had a better way to track this kind of credit than these ways, and I would be very open to an overhaul of how academics give and receive credit.  But until then we need to act ethically in the environment we do live in.

If they wanted co-authorship they should have published sooner.  The fear of getting scooped is what keeps us productive. This would provide a perverse incentive to collect data and never publish it.

These are not ethical arguments. They boil down to: “their sloth justifies my theft.”

But taking on potentially hostile co-authors is not a good idea. Forced collaboration is a terrible idea.

I absolutely agree!

(And let’s put aside the question of why this person would be hostile towards you, and how you’re sure you’re in the right.  After all, as I discussed in Part I, being allowed to do something doesn’t mean you’re not being a jerk for doing it. But let’s assume arguendo you’re in the clear and they’re hostile for some other reason than your misbehavior.)

Here’s what I think the radical part of my suggestion is based on:

co-authorship does not have to mean collaboration

The minimal rights of co-authors are actually set out in the AAS Ethics statement:

All collaborators share responsibility for any paper they coauthor, and every coauthor should have the opportunity to review a manuscript before its submission. It is the responsibility of the first author to ensure these.…All authors are responsible for providing prompt corrections or retractions if errors are found in published works with the first author bearing primary responsibility.

See? No real collaboration beyond the opportunity to review a manuscript. If Candice, Amber, or Brie (all of whom have been offered co-authorship) make demands on the paper that Joe’s team disagrees with, Joe has every right to say “no” and the proposers have every right to stay off of the paper.

But that’s not really a choice. If these teams don’t want to collaborate, then the proposing team shouldn’t be on a paper where they did not get a say in the methods and conclusions. They might even disagree with the conclusions! And if they make a principled stand and decline to be on a paper they disagree with, they don’t get the credit they deserve.

This is true, but this is not a problem with my proposal: it’s a problem with the concept of co-authorship in general, and it comes up all the time. Many co-authors do not agree with papers or in some cases do not even read papers they are on. Regardless of how severe a problem you think this is with our current model, it is not an excuse to keep proposing teams off of your paper.

But it’s also not a general solution: ethically people must refuse authorship if they disagree with a paper. As co-authors they would be “responsible” for it, after all.

Because this is a general problem, and not an objection to my proposal per se, I offer my general solution: I favor requesting that every author provide a one-sentence description of their contribution to the paper. If an author is only on the paper because they took the data, they should state exactly that.

So if an author disagrees with the content of the paper they can add that in, too (it would be reasonable to limit such qualifications to, say, 140 characters in most cases; a bit more if necessary). That way everyone’s contribution and responsibility for the result is clear and unambiguous, and credit lands where it is due. I have done this several times, even though there were no contentious issues to hash out.  In this way authors can state exactly what their responsibility for a paper is, if they like.

I still think it’s wrong to bring on co-authors from competing teams that didn’t even contribute to the text of a paper!

I don’t think this is really at the emotional core of objections to my proposal.

Many of us have had to deal with that that one senior team member that totally slacked off and didn’t even send in comments and may not have even read the manuscript. They probably don’t really deserve to be a co-author, but we still include them with little more than a tinge of annoyance because that’s the community norm: you invited them on at the beginning, and you should presume that they read the manuscript and were happy with it and had nothing to add, and it would be rude and awkward to take them off. Yes, sticklers should insist will that they contribute or take their name off, but this situation does not arouse the sort of reflexive opposition that my proposal does.

Whereas the thought of adding members of a competing team as similarly “silent” co-authors makes us uncomfortable, even tough they unequivocally contributed much more than the slacker to the science and an equal amount to the manuscript.

Why do we feel so differently about these situations? Not because the proposing team is less deserving of authorship than the slacker, clearly. It’s partly because they are “the competition” perhaps, but mostly, I think, because it’s the community norm that we don’t invite strangers onto our papers.

I assert that this norm is unethical and we should change it.

In the next part: some practical issues and final thoughts, including a skeleton data policy proposal for MAST (for Josh).

Who Should Be an Author on a Paper? II: Credit as Currency

In Part I I argued that if you use other peoples’ data in your own paper, you should offer them co-authorship on your paper.  In this part, let me make flesh out the theory behind my proposal, in particular why the policy exists, so that we can apply it where appropriate.

I had a math professor in college who made an analogy that has stuck with my all my career: the product of the Academy is ideas and research output and the currency we use to trade in this product is credit.  We cite, we co-author, we acknowledge. This is at the heart of the AAS Ethics Statement’s rule: if someone did work that made your paper possible, you pay them back with credit in the form of a co-authorship.

Now, the policy is clearly too broad. Sometimes the appropriate currency is a citation, not co-authorship.  In particular, if data have already been published then the norm in our profession is that you don’t need to include them as a co-author; you can just cite the publication.

In many cases, successful proposals are citable and appear on ADS. This provides another way to give credit for using other people’s data, although it is imperfect because proposals are rarely cited, so it’s not really a good way to accrue credit. It’s not a currency that is generally recognized by, say, promotion and tenure committees. If we could change that (make the citations worth more and make them common) it would solve the problem, but that seems more radical to me than my proposal.

Also, the AAS Policy does not define its scope. Which enablers of science deserve authorship?  The AAS guidelines are no help here.  What about the armies of PhD astronomers at STScI and IPAC that enable and reduce NASA space telescope data? The engineers who built the telescopes? The telescope operators? The staff that cleans the dorm rooms at the observatories?

There are professional norms here, but it’s surprisingly hard to articulate them. Note that I’m not defending those norms, just trying to figure out exactly what they are.

Going back to the currency analogy helps a bit here: in the norms of our profession, who needs and appreciates citations and co-authorship as professional currency that advances their careers? Not the cleaning and cooking staff at the dormitories. Many telescope operators do not, but many telescope staff astronomers do.  Many people who write data pipelines and archiving software do.  Certainly instrument designers and builders to, as do some members of the shops that construct the instruments. An imperfect shorthand for this might be “anyone eligible for membership in the AAS” (or their country’s equivalent).

Here I think there is an ethical obligation on observatories and science centers that produce data to offer guidance to users on how their staff that accrues and values citations to get them. This means that data pipelines and instruments need to have papers that can be cited, and staff astronomers that assist with observations in any way need a clear path to getting credit for the science they enable. These centers also need to communicate with their users about what these policies are and what appropriate citation and authorship practices for their employees entail.

OK, having laid the groundwork here, in Part III I’ll defend my proposal from Part I.

Who Should Be an Author on a Paper? I: AAS Ethics policy

I started a really long Twitter conversation by blurting out a radical-sounding assertion that I’ve been mulling over privately for a long time.  This series of posts is an attempt to justify my (apparently) rather unpopular position.

The AAS Ethics Policy States:

All persons who have made significant contributions to a work intended for publication should be offered the opportunity to be listed as authors. This includes all those who have contributed significantly to the inception, design, execution, or interpretation of the research to be reported.

Sounds reasonable! But—like a lot of ethical maxims—this apparently banal statement can be tricky to apply in practice. I actually like this rule a lot and think it should stand with only minor clarifications, but I assert that it is strongly inconsistent with the norms of our profession.  We could change the norms, or we could change the policy.  I think a compromise is in order.

Let’s look at a concrete example:

Consider a team of researchers that proposes for some Hubble Space Telescope time. After submitting the proposal but before analysis of the data begin, two of the co-I team members (let’s call them Amber and Brie) that worked hard on the proposal leave the group. Amber gets a job in industry, and Brie gets a faculty job elsewhere and has no time to work on the project any longer.

The PI (let’s call her Candice) and the rest of the team get the data and write a paper. Do they have an ethical obligation to include Amber and Brie on the paper? I think it’s clear that they do: they clearly “contributed significantly to the inception [and] design…of the research to be reported.”

OK, that’s an easy one. Now let’s make a minor tweak.  Let’s say this was a DDT proposal. That means that the data have no proprietary period, and go public as soon as they are ready. Simultaneous to the above events, another team (led by Joe) downloads the very interesting and useful data, analyzes it, and prepares their own paper.  Is Joe ethically obliged to offer Amber and Brie co-authorship on the paper?

Our professional norms say “no”: this is a different team using public data; why should Amber and Brie be involved?

But our professional society apparently says “yes”: by the book, this situation is no different than the first one.  Amber and Brie “contributed significantly to the inception [and] design…of the research to be reported.” Full stop. By this rule, the entire proposal team should be on Joe’s paper!

In fact, the AAS policy has things exactly backwards of our professional norms: many astronomers would, I think, consider Joe a bit of a jerk for scooping Candice. Even though he’s allowed by NASA to publish the data, there is a general etiquette that we don’t do that, or at least that we ask first, or at the very least an understanding that Candice is perfectly justified being upset about it. But there is also a broad consensus that Joe doesn’t owe Candice co-authorship.

So the AAS Policy is clearly out of step with our norms. Should we change the policy?

I actually think not. I agree that Joe’s act is poor form, but allowed; my (apparently radical) proposal is that Joe should seriously consider inviting Amber, Brie, Candice, and the entire proposing team to be co-authors on his team’s paper.

In Part II I’ll flesh this out.

The PITS

There is zero evidence for ancient aliens in the Solar System.

OK, now that that’s out of the way…

Sooooo, I wrote a paper and it’s been accepted to the International Journal of Astrobiology. Yay!  Astrobiology refused to have it refereed, claiming it was out of scope, which I admit made me grumpy:

But that’s fine; if they don’t want solar system artifact SETI in their journal, that’s their loss. Perhaps they’ll come around as Breakthrough Listen starts its survey of Solar System objects for radio emission. Anyway, that’s all water under the bridge now.

Normally I would have done a big roll out, a 10-part slow blog of the whole saga, and describe the paper in detail but…

  • I was traveling from California to the Astrobiology Science Conference near Phoenix when I learned it was accepted, so I didn’t have time to blog it.
  • I wanted to get the preprint out right away, during AbSciCon, since it’s my first astrobiology paper, and I thought having it hit the arXiv during the conference would make for good conversations. Also, Breakthrough Discuss had just finished, so SETI was also on people’s minds.
  • A major family emergency had just struck (everyone’s fine now), and I had no time to blog, or even do much of anything at AbSciCon (I have a long draft of a blog post almost ready to go that I haven’t had time to even look at in two weeks).
  • I think the paper is very short and readable—an easy register, not too much jargon—so if you’re interested in what’s in it I recommend you just read it.  My blog would just quote from it, for the most part, anyway.
  • I thought I’d do a slow-blog later—I wasn’t really expecting much in the way of press to scoop me; it’s kind of a fluffy paper (to use Steinn’s term for it)
  • That said, I had shown the paper to the great folks at the Atlantic science desk (Ross Andersen asked what I had tweeted about above) and so I knew it would be treated well if it got any press at all.

Well, it certainly got some press! Not #TabbysStar #AlienMegastructures levels of press, but enough that I have a very busy week!

The Atlantic article was nice, and if that had been the main source of news stories, I think it allwould have gone much better.  But somehow, the yellow press found the paper on their own on the arXiv (do they read astro-ph daily?!) and ran away with it without asking me what it was about. The Daily Mail, that British rag of a tabloid, claimed that I “believe[] the aliens either lived on Earth, Venus or Mars billions of years ago.”

Wow. Things went downhill from there as the NY Post, repeated the article, and USA Today posted a video that was even worse. The only saving grace is that according to the worst of the articles, the irresponsible astronomer posting Ancient Aliens papers on the arXiv wasn’t me:

Gizmodo got to talk to me after the craziness began, and they were great and helping me to reframe things more appropriately.  Universe Today was really careful to get things right, too, as was NBC Mach.

For the record, the premise of my paper is the fact that we have no evidence for any prior technological species in the Solar System. My paper asks, is this a dispositive null result? That is, has our paleontology on Earth and mapping of the larger Solar System bodies basically proven that we are the first such species around?  After all, the idea that we are not is very old (read the paper—the earliest citation I got from folks contributing to my Twitter research was 1900 years ago!).  This was actually an area of active discussion in astronomy until the advent of robotic exploration showed no canals on Mars, no ruins of cities on Venus.

But is it too soon to rule out the possibility entirely? I thought the idea needed to be formalized, to have a name, because it seemed to me the literature had forgotten about it prematurely. Papers on searches for alien artifacts in the Solar System always seem to implicitly assume such artifacts would have come from an interstellar species—but if Venus was ever inhabited, couldn’t its inhabitants have something to find?

So, I ask, what’s left?  Ancient things are hard to find, because planetary surfaces erode and subduct things away. We have a pretty good understanding of life on Earth, and the window between “so old we wouldn’t know about it” and “so recent we couldn’t have missed it” might be very narrow. But is it really closed?

I don’t know, but it seems like the kind of question we have the ability to answer today. Someone should answer it! How long would free-floating artifacts in the Solar System last? How far beneath the surface of Mars would technology have to be to survive billions of years? And how deep can we probe with radar? How long ago could Mars have been inhabited?  Venus How wide is the window between technology we could never discover because it has been too long, and technology we know isn’t there because we’ve checked?

That’s the conversation I wanted to have.  That’s why I wrote a paper on Prior Indigenous Technological Species: not because I think they exist, but because we’re at the point where it should be possible to say for sure that certain types of them didn’t. The end of the paper is all about the things we can do to start drawing some conclusions.

And that’s a neat SETI (SPITS?) project someone should undertake.

At least I think it’s neat.  Your mileage may vary. In that vein, let me use this as an opportunity to address a weird misconception that the SETI grumps in astronomy have. Apparently, the only reason to do artifact SETI (or even mention it in a paper on another topic) is to get attention. Seriously, I’ve had good astronomers I respect make this claim and defend it when challenged. It’s a real attitude out there.

Well, the truth is exactly the opposite. When trying to do artifact SETI, I have inevitably caught all the wrong kind of attention from the yellow press and the ufologists.

And it is mortifying.

So why do I carry on? Certainly not for the attention. I carry on because it’s interesting, and because lots of other colleagues I respect tell me they find it fascinating and worth working on.

Now, I’m not claiming to be some sort of martyr for the cause, here. My point is that it’s a problem worth working on despite the attention, not because of it, and so the SETI grumps that think otherwise should seriously reconsider their assessment of the motives of SETI researchers.

Now excuse me while I answer all these emails from Coast to Coast and ufologists sending me pictures of clouds.

 

On Outré Ideas

David Stevenson has a nice Commentary in Physics Today:
http://physicstoday.scitation.org/doi/full/10.1063/PT.3.3507

He argues in defense of “crazy” ideas in science. He categorized three kind of “crazy”:

The First Kind is simple crackpottery: people who don’t know enough science to articulate a real scientific idea, and do not understand its interconnectedness well enough to distinguish outré ideas from nonsense ones. He says this is the most common and least interesting (and he’s right, except as a study in sociology)

The Second Kind is when good scientists take a fresh, naïve look at a new field. He writes:

Inevitably, such excursions can look like the actions of a dilettante…One is then accused of speculation. I occasionally sense from colleagues some disdain for scientific speculation, perhaps because it is cheap: It seems to require relatively little effort and commitment.

He argues that hard, serious speculation is rare and important. This is certainly something I’ve tried to engage in. My excursions into lunar geology theory, the Faint Young Sun problem, and even SETI (at first) were certainly in this category. Indeed, the reactions I got to our lunar highlands work from the lunar science community ranged from the pleasant to the snide (Dave himself was polite, but dismissive—though, after reading this I wonder if I misread him.  Caltech GPS was certainly the most receptive audience I found.).

Physicists are prone to this sort of work (consider Richard Muller’s forays into climate science (and borderline denialism)) to the point of cliché (one of my favorite SMBC comics), and SETI seems to be a favorite destination for dilettantes from all fields.

The Third Kind are when established leaders in their fields upset the table with entirely new perspectives.  One occasionally sees cals for such ideas: Lindy Elkins-Tanton acknowledged the need for new ideas in lunar formation theory in Nature, and Michael Inzlicht in psychology has done serious soul searching, wondering if his career, indeed much of his field, is based on bad statistics. But these ideas are not always invited or even welcome; Stevenson’s examples are Hoyle’s Black Cloud and steady-state cosmology, and the idea of emergent gravity.  He finishes with a quote from Neils Bohr to Wolfgang Pauli: “We are all agreed that your theory is crazy. The question which divides us is whether it is crazy enough to have a chance of being correct.”

Stevenson briefly mentions a “portfolio” of ideas, and this brings me to one of my favorite papers, Avi Loeb’s banquet lecture (here at Penn State!) about how young scientists should divide their research time. He argues for some fraction of time to be spent on “venture capital”, analogous to spending some of your financial portfolio on high-risk, high-reward investments. (Avi may have been inspired by Eric Weinstein’s lecture on the topic, h/t Michael Neilson for pointing me to it) Avi acknowledges that there is actually a lot of acceptance for work on outré topics, but argues that the natural conservatism of the Academy and science tends to favor no more than 5% of one’s efforts there.  He argues it should be more like 20%.

I think between 5-20% is right, in an average sense. So some scientists should spend 100% of their effort on safe “bonds”, others a lot more on venture capital (my last few years have involved much more SETI than I had planned for), but if as a whole we’re working between 5-20% of the time on Second and Third Kind speculation, I think we’ll do well.

I use “outré” above instead of “crazy” (and I put the latter in scare quotes) because the latter is a slur for people with mental health issues. I appreciate it’s often not meant that way, but I think society is slowly realizing how common mental illness is, how badly and unjustly it is stigmatized, and how casual uses of terms like “crazy” to mean “unexpected” or “weirdly different” reinforce that stigma.  Also, “crazy” isn’t even the right word here—by using it, Dave is facetiously implying that one would have to be mentally ill to have come up with the idea, but all he needs to make his point is to say that it’s nonobvious, very clever, and outside the normal thought process—outré.

Ethics of Minor Rules

Following up on the ethics in basketball thing:

I was arguing with my neighbor about the poor aesthetics of a sport where breaking certain rules is encouraged and accepted (as I wrote, I think basketball is broken), and he argued that there aren’t any “rules” in sports, really, just tradeoffs: if the penalty is worth it (or the whole point of it) it’s not unethical to break the rule, it’s just part of the game.

The “don’t intentionally foul” article helped crystalize this for me. There, Lopez argues that there are regulatory rules and “constitutive” rules: that the latter define the sport, and the former are essentially arbitrary details specific to a particular variety of the game (pro vs. collegiate, pickup vs. formal, etc.)  I had sort of a similar feeling, but my argument last time helped define things better for me.

The constitutive-regulatory spectrum has no sharp boundary: enough regulatory rules and at some point it’s a new game (witness rugby, soccer, and the many versions of football; all of them evolved from the same sport).  The real distinction for ethical purposes is which rules the community expects to be followed—what’s “normal”.

Consider an analogy to law. Is it unethical to break the law? Remove for a moment the situational dependence involved in, say, civil disobedience or justified homocide, and just consider breaking a law because you want to or it’s convenient. In this case, it depends on how society views the law.

For instance, if it’s the middle of the night and there is zero traffic, is it ethical to jaywalk, or should you dutifully go to the corner and push the walk button and wait for the light? Clearly, jaywalking is no big deal; the ethical consequences of jaywalking in that situation are probably not worth the thought given to the problem in the first place.

If the law is that the speed limit is 65 and you’re doing 67, this is not a big deal and generally not unethical, per se.  If you’re doing 65 and you’re still the slowest one on the road (it happens!) there is a good argument that you should speed up.

Don’t buy it? Look at how outraged (and actually dangerous) people became when some students pranked a highway by driving the speed limit, in formation:

This is not the cleanest example—the law says you should leave the passing lane clear for faster traffic to move ahead—but it more than illustrates what every driver knows: the speed limit is not a hard limit, and no one expects it to be universally honored. Murder, of course, sits at another end of the spectrum, because laws against arbitrarily killing people are pretty fundamental to the functioning of society; jaywalking is more of an arbitrary detail that can often be ignored.

Back to sports: I think “constitutive” rules in sports aren’t more important because they’re foundational per se, but because being foundational they’re less likely to have a consensus that it’s OK to break them. The “Mano de Dios” soccer goal is notorious because the rule that was broken—don’t use your hands in soccer—is foundational (“constitutive”), which makes the violation seem especially flagrant. Compare this to the equally legendary 1999 World Cup shootout:

The US won in part because the US keeper, Scurry, left her line before the ball was kicked, which is a technical infraction (go to 0:43, and compare Scurry’s jump on the kick to the action of Chinese keeper, Gao, holding her line until the moment of the kick at 1:00).  This isn’t to minimize or question the validity of the US victory: this technical infraction is so minor and violated so routinely that even most soccer fans aren’t even aware it exists.  My point is that the difference in reaction to these two goals shows that Lopez’s “constitutive” vs. regulatory continuum maps pretty well to my “community norms” test.  And the latter is better connected to ethical reasoning.

Why is socially acceptable (even expected) to break some rules and not others outside of sports? That’s a deep topic, but my point isn’t to explain why, just to show that it clearly is, because it’s a key element in any ethical analysis of when it’s OK to break a rule, whether it’s a speed limit or a basketball foul or a high crime or misdemeanor. It’s not a dispositive test (society can condone evil law-breaking behavior, like lynching) but I think it’s the difference between my conclusion and Lopez’s.

Ethics in Baskeball

According to this Ask the Ethicist column, it’s unethical to follow a basketball coach’s order to strategically, intentionally foul?!

This is a neat, regular column written by Rock Ethics Fellows here at Penn State, and I normally find them instructive, but as a Rock Ethics Fellow myself I find this one to be totally wrong.

You can’t derive specific ethical rules purely from first principles, as this author tries to do. Ethics are normative, and the norms are set by the expectations of the community. That’s what “unwritten rules” are all about!  It is widely accepted in basketball that an intentional foul is a legitimate strategy in basketball. This analysis is otherwise good, but it comes to the wrong conclusion because it ignores this essential element of the problem.

Instead, the author here tries to draw a line between constitutive and regulatory rules to come to his conclusion, but this line is arbitrary and subjective, so it leads to a squishy answer. He argues that free-throws, for instance, are not intrinsic to the game of basketball, which would be news to most players! He also appeals to the NCAA’s stated core values (HA!) to argue that intentional fouls violate those values, because strategic fouling is only appropriate when “victory is regarded as the main goal”, but “victory is not mentioned as the main goal of college sports”.  That may be true for the endeavor as a whole (after all, there are just as many winners as losers!), but winning is certainly the main goal of an individual game. By definition! If that’s not in the NCAA rules, it may be because it’s so obvious, so fundamental to the definition of an athletic competition, that it doesn’t need stating.

This is not to say that following community expectations is a defense against immoral behavior—sometimes the community consensus is unethical. But there’s no great social ill in a sport having a too-lenient penalty for purely technical infractions like this (intentional fouls are almost always highly technical infractions, not attempts to injure or be unsportsmanlike).

And it’s true that cheating to win is unethical, especially in college sports where sportsmanship is supposed to be a core value.  But by this argument I could try to say that head fakes are unethical, because I am deceiving my opponent in an attempt to win the game. Of course, head fakes are obviously not unethical, and the reason isn’t that head fakes aren’t disallowed by the rules (lots of forms of cheating aren’t mentioned in the rules!), it’s because they are considered a legitimate strategy, just as intentional fouls are.  That is, the line is between what’s considered “fair play” and what’s not is set by community expectations, and we’re back to the unwritten rules that this author left out of his analysis.

Now, I happen to agree that the rules of basketball are broken, in that if it is regularly to your advantage to break the rules then the penalty for rules-breaking is too light. But that’s an aesthetic concern, not an ethical one.

I can think of lots of concerns that would make it ethical to not try your best to win a game. There are lots of heartwarming stories of athletes refusing victory in the name of good sportsmanship. But a quixotic devotion to avoiding fouls is not one of these higher values—your opponents will feel confused and robbed of a fair victory, not impressed by your ethics.

[Edit: John Gizis points out that the author could be referring not to end-of-game fouls to stop the clock, but to “Hack-a-Shaq” strategies.  In that case, the author is probably right. This strategy is not universally accepted, even in professional basketball, so in collegiate sports (with its higher emphasis on character building and sportsmanship) it is probably outside of community norms.

That said, it’s hard to really distinguish “Hack-a-Shaq” from the intentional walk in baseball.  I wonder if  Francisco Javier Lopez (the author of the ethicist column) would say that’s unethical too?  I presume so, although I think it’s fine (because collegiate baseball players expect it and they think it’s fine).

Second edit: Chris Palma points out that the intentional walk in baseball has a similar ethical ambiguity. A situationally strategic walk (to put someone on first to create a force-out situation when runners are already in scoring position, for instance, or pitching around the 8 spot to get to the pitcher with 2 outs) is, as far as I know, universally accepted strategy, and so (I would argue) not unethical.

But used as the baseball equivalent of “Hack-a-Shaq”—routinely pitching around dangerous hitters—was somewhat controversial when it happened to Bonds, Sosa, and McGuire. I could imagine that in collegiate sports a similar approach might be considered unsportsmanlike.

So it’s all a matter of degree.  For instance, is a defensive player with a “five fouls to give” mentality playing unethically? After all, anticipating that you’ll play so aggressively that you foul your opponents five times by the end of the game isn’t much different from Hack-a-Shaq in principle. Again, I think you’d have to ask the players and coaches, not try to deduce it from first principles.

Third edit: More here in part II.]

Przybylski’s Star V: Origins of the Idea

Part V/III.  Part I is here.

The earliest reference I cited to the idea of artificial species in stars was from Howard Bond, who said he thought Shklovski & Sagan dealt with it in their book.  I finally got my hands on a an early edition.

It looks like Frank Drake and Iosif Shklovskii were first with this idea.  From the 1966 edition of Shklovskii & Sagan’s “Intelligent Life in the Universe”:

…we should mention independent suggestions by Drake and Shklovskii that, if not the communication of large amounts of information, at least the communication of the presence of a technical civilization, can be effected through the use of markers.  Drake and Shklovskii envision the dumping of a short-lived isotope—one which would not be ordinarily expected in the local stellar spectrum—into the atmosphere of the star. In any case, the material of the marker should be of a type that is difficult to explain, except as a result of intelligent activity.

They conclude with a sentiment I tried to articulate at the end of Part IV.

Remarkably enough, the spectral lines of one short-lived isotope, technetium, have in fact been found in stellar spectra. Its half-life is around 2×105 years.  However, technetium lines have not been found in stars of solar spectral type, but rather oly in peculiar stars known as S stars.  In fact, as we saw in Chapter 8, the discovery of technetium in the S stars have been used as an argument for contemporary stellar nucleosynthesis.  This example illustrates on of the difficulties with such a marker announcement of the presense of a technical civilization.  We must know a great deal more than we do about both normal and peculiar stellar spectra before we can reasonably conclude that the presence of an unusual atom in a stellar spectrum is a sign of extraterrestrial intelligence.

I see this as a feature of artifact SETI: anomalies are inherently interesting.

Changes to AAS Governance

The American Astronomical Society is about to change its governance structure for the first time in 50 years in response to a task force report on the topic. Here is the task force chair’s description of the process, and here are the executive officer’s thoughts.

Here is a summary of the changes (from the executive summary):

To improve responsiveness and assure timely action on important matters, and to move toward a model consistent with best practice, we replace the AAS Council with a Board of eleven members that meets monthly. To assure effective communication between this Board and the entities in which our members work to advance the interests of the Society and our profession — our Committees and Divisions — we provide a Board liaison for each. Most importantly, each Committee and Division Chair is scheduled to attend from two to four Board meetings per year in which their issues are pre-ordained agenda items.

To enhance inclusivity in the governance of the Society, Committee members and chairs will no longer be appointed by the Board (nee’ Council); members will be derived from volunteers and selected by the existing committees, and the committee members will, in most cases, elect their own Chairs.

To further involve a broader community in setting the strategic directions of the Society, the AAS Council will be reconstituted as a body — the Strategic Assembly — including the Board and the Committee Chairs of the eleven Standing Committees as well as Division representation. The Assembly will meet twice a year at the Society’s scientific meetings 1) to foster collaboration among committees and with the Board, 2) to improve communication, and 3) to guide the strategic thinking of the Society.

The many changes, large and small, outlined in the attached document have been carefully designed to promote inclusivity, foster communication, embrace creativity, and maximize transparency — in short, to enhance the functioning of the Society so that it will be the welcoming and natural home for the many people who, in a variety of different capacities throughout their careers, work to enhance and share humanity’s scientific understanding of the Universe.

Full details in the task force report.

The AAS would like feedback from its members specific to the drafting of the bylaws to be sent to the task force chair, David Helfand, and general comments can be emailed to AAS President Christine Jones.

Comments can also be sent anonymously here.  Comments will be appreciated before May 12.

There will be a town hall about these governance changes on 7 June at 12:45pm at the summer AAS meeting

When to Violate Confidentiality of Peer Review

Peer-review is an important part of scientific communication. It is not a panacea and is quite fallible, but it serves several functions well:

  1. It gives journals a “filter” against unprofessional work (including crankery and its cousins)
  2. It gives authors and editors the opportunity to receive unfiltered feedback on how a paper will be received. The “unfiltered” part is ensured in part by the anonymity of the referee (which the referee can waive).
  3. It provides a fresh set of eyes from an expert who can help spot errors and mistakes
  4. Ideally, it gives authors constructive feedback to improve their work
  5. It does all of this confidentially, so when science is finally presented to the world it has already been vetted and mistakes have been fixed (this step is optional for authors—they can post preprints whenever they like).

Refereeing papers is an unpaid service to the community that scientists provide. I think of it like jury duty: I rely on good referees to fairly judge and improve my papers, and so I try to write at least as many referee reports as I receive, and be the sort of referee I hope I’ll get on my papers (I have to referee more than I am refereed because some scientists are very bad referees (slow, delinquent, unduly harsh, sloppy, etc.) and the rest of us need to pick up their slack.

The confidentiality of the process is important to the journals, and to some authors and referees. For the authors, this means they do not need to worry that their work will be scooped, that their mistakes have a chance to be fixed before they are made public, and that they have a chance to respond to any unfair criticisms in a report. For journals, this confidential review process is one of the primary values they add to papers (some of the others being editing, typesetting, publicity, and curation). For referees, it means they can frankly help improve the literature without being concerned about the reaction of the paper’s authors.

There are other ways to do this; there are good arguments for entirely open peer review, as well.

One interesting issue is whether the referee should be bound by confidentiality. There are good reasons why a referee might feel they should publish their review. This is especially true in instances where there is evidence of fraud, for instance. The case is made here:

The argument is that there can be good reasons to publish your report (as open peer review shows; read the article), and because the referee holds the copyright to the review.

But in the AAS astronomy journals, confidentiality of referees report is explicitly part of the journal ethics statement that binds referees and authors. Referees thus promise to keep the reports confidential.

Of course, this could lead to an ethical dilemma if publishing the referees report would improve science. But I think there are many perfectly ethical routes one can take to avoid that dilemma:

  1. Polish the relevant parts of your referee’s report, make sure it only refers to the published paper, and not the (strictly confidential) manuscript, and publish that on the arXiv as a “comment”.
  2. If the content of the manuscript needs to be mentioned, go directly to the editor. The purpose of confidentiality is to preserve scientific integrity; the editor has the authority to waive confidentiality, or take other actions to rectify the problem.

For instance, a climate science journal Environmental Science Letters rejected a paper by a climate skeptic, Lennart Bengtsson, on the basis of two very unfavorable referees reports that heavily criticized the science. The reports even offered suggestions for how to improve the paper and make it publishable.

Prof. Bengtsson cherry-picked a few words from one of the reports and (violating confidentiality) complained to the media that the paper had been rejected because of its conclusions for political reasons.

The journal, in this case published a rebuttal that included the entire referees reports, exposing the mendacity of Prof. Bengtsson’s accusations. Review confidentiality is at the discretion of the editors, so they can (and should) waive it when it is not serving its purpose (preserving scientific integrity).

If, for some reason, the editors were to enforce the confidentiality to the detriment of scientific integrity, then a referee may need to violate the rules and publish their review.  In this case, I recommend checking (confidentially) with others familiar with the issues, especially an ethicist, to make sure you are in the right.  After you post your report, the journals may complain, and you may face repercussions from your professional society, but if on balance you are behaving ethically you will have a good defense.

Astronomy and “Meta-Astronomy”: An Allegory

My blog is about astronomy and “meta-astronomy.” By the latter I mean the stuff that isn’t strictly astronomy research but is necessary for, or relevant to, its practice. I think both astronomy and meta-astronomy are appropriate topics for journals, talks, conferences, blogs, and research, especially since the line between them is not sharp. Here is an allegory I thought of this morning to illustrate this point.


This story is pure fiction. The allegory is purposeful, but literal similarities to real people, events, and fields of study—while perhaps not entirely coincidental—are not intentional or relevant.

Joe is an exoplanet astronomer at Research Center. Like many exoplanetary astronomers, his PhD thesis was on planetary science, which gave him a firm foundation for studying exoplanets, their composition, internal structure, and their potential habitability.

He goes to an internal departmental lunch talk by Diamond. Diamond is another exoplanet astronomer at Research Center. Like many exoplanetary astronomers, Diamond’s PhD thesis was in stellar astronomy. Her talk is about the complications of deriving exoplanetary properties from observations: starspots, convection models, mixing length assumptions, and photometric errors.

On the way back to his office, Joe and another colleague with a planetary science background, Jack, banter about how it feels to attend these “stars” talks: they agree that they appreciate stellar astronomy is important for their work, and that they should know that stuff better than they do, but they’re glad to get back to their offices so they can focus on their part of the science.

Joe is organizing a conference on exoplanets focusing on exoplanets’ composition, internal structure, and habitability. It’s going to be an important meeting where exciting new results will be presented and discussed for the first time, and he wants the best exoplanet astronomers to attend so it will be maximally successful.

He heads over to Diamond’s office and asks if she will please attend, since she is one of the best exoplanet scientists. She points out that all of the planned sessions titles contain geology jargon, and there don’t seem to be any talks about stars. Perhaps she could recommend some speakers for a panel on that?

Joe explains that the success of the conference will depend on it staying focused on its primary topic, and he doesn’t want to “dilute” the science with “stars stuff.” She points out that “stars stuff” is actually very relevant to exoplanets. Joe hastily agrees, but reemphasizes that he wants *this* conference to stay focused on geophysical aspects of planets. Will she please attend?

Diamond points out that she has limited time and can’t accept every conference invitation. She says it looks like Joe is giving only lip service to the relevance of stellar astronomy to his work, and that his actions indicate he doesn’t really think it’s very important at all. She points out that the “focus” of his conference excludes much of what she spends much of her professional time thinking about. Disappointed, Joe says thanks and goes back to his office.

He is annoyed. He would like to work more closely with Diamond, since they are both exoplanet astronomers in the same department, and they generally get along well, but it seems like she wants to drag stars into every discussion. He wishes she could compartmentalize better.

Diamond is annoyed. Joe professes to want to work with her, but can’t seem to appreciate that stars are integral to her work. She finds many planetary scientists are like that: they seem to imagine that they can divorce planets from the stars they orbit. Of course, that’s true in a purely theoretical, pedantic sense: many planetary interiors are more-or-less insensitive to the most of the precise properties of the star they orbit. But in practice you can’t measure anything about those planets without observing the stars and knowing their properties. She finds Joe’s myopia on this point frustrating.

Back in his office, Joe is glad to be able to get back to work on his geophysics. He looks at his panels: his science organizing committee has done a good job of getting many of the best speakers and presenters, even though many people had to decline.

Diamond looks at the preliminary program of Joe’s conference. All of the panel speakers come from a planetary science background. She finds a single stellar astronomer on the registration list. She thinks about going to Joe’s office to point this out, but she’s had this conversation with him before. He’ll say that he tried to get stellar astronomers to attend—he even tried to get her to attend! But, he’ll say, they all declined, so what does she want him to do?


I hope my astronomer friends will not be like Joe. When a fellow astronomer tells you a subject you find to be “meta” is important to their work and needs to be part of your science, your conference, or your work, accept that, embrace that, and act on that. Don’t nod and then go on treating it as a tangent or a distraction to your work. And when you look at the composition of panels and find homogeneity, don’t let “they all declined” be an excuse. Ask “what is it about my panel that made only certain kinds of scientists end up on it?”

Przybylski’s Star IV: Or…

Part IV of III.  Part I is here.

A coda: Howard Bond correctly points out that my three explanations are only necessary if a very plausible and less interesting explanation is wrong (a caveat that I had in an early draft of my posts but edited out unintentionally.)

The identification of short-lived actinides could be a mistake! The Gopka et al. identification of these lines was made in a journal I had not heard of, Kinematics and Physics of Celestial Bodies, apparently originally in Russian. As far as I can tell, the paper has been cited exactly once, by the Dzuba et al. paper that proposed the metastable heavy isotope.

The journal and language of the Gopka et al. paper aren’t necessarily problems, of course, but they do raise eyebrows. The fact that it has not been cited could mean that the paper was simply not read (not surprising, given the journal), or that everyone who studies the star that saw the paper decided it was not worth citing, even to refute it.

[Edit: Steinn is much better at this than I am.  He points me to a 2003 AAS abstract by Crowley et al. supporting the existence of short-lived isotopes, a topic Howard Bond also mentioned on Facebook to me. Steinn also finds this paper and this one which I think I missed because I didn’t realize that promethium, a lanthanide, has no isotopes with half-lives longer than 20 years.

The Mkrtichian paper I linked to in the last post mentions Bidelman et al. PASPC, 336, 309, as supporting the short-lived isotope interpretation, and conference proceedings by Yushchenko, Gopka, & Goriely that ADS doesn’t know. Goriely discusses mechanisms here.

So the claim is stronger than I originally hedged in this post.  It’s put best in this followup paper by Crowley it al., originally shown to be by Brian Davis (but which I only just found again, now that I’m thinking of Pm): “The spectroscopic evidence is strong enough that we would declare promethium to be present without hesitation, if any of its isotopes were stable.”  In their other words, it’s only the strong prior against finding unstable isotopes that makes them hedge.]

The mystery of Przybylski’s Star is still a very good one if there are no short-lived actinides isotopes in the spectrum—the identification of the stable lanthanides seems quite secure and fascinating and it remains the most peculiar of the peculiar A stars—but it would mean that it is much more plausible that technical but mundane explanations for the star exist.

[P.P.S. There is now a part V/III about prior art by Drake and Shklovskii.]