A group of physicists has retracted their preliminary report in the GCN Circular of a massive star-sized explosion after deciding that what they’d really observed was another phenomenon.
Although we could try to explain this, we’d rather leave it up to Giacomo Vianello, an experimental physicist at Stanford University, who was a member of the research team.
Vianello told us:
on March 30th 2014 the Large Area Telescope, orbiting onboard the Fermi NASA spacecraft 560 km above the Earth, detected high-energy gamma-ray emission from a region of the sky close to the position of a Gamma-Ray Burst (GRB 140330A), previously detected and broadly localized by another instrument onboard Fermi (the Gamma-ray Burst Monitor, GBM). GRBs are the largest stellar-class explosions in the universe, releasing in time scale of the minute energies larger than what the Sun will emit in its whole life. Discovered in the sixties, they are now routinely observed (1-3 per week). We therefore issued a circular:
http://gcn.gsfc.nasa.gov/gcn/gcn3/16048.gcn3
claiming the detection of the GRB. This is a standard procedure. We detect 1-2 GRB per month with the Large Area Telescope.
A few hours later, when we received more data downlinked from Fermi, we realized that what we thought to be the high-energy counterpart of the GRB did not look at all like it. While GRB counterparts decay in luminosity quite fast, this source was steady. Looking back in the data previously collected by Fermi since its launch in 2008 we realized that such source has been active sporadically over the whole mission span, and therefore is unrelated to the GRB (which just happened by chance to go off close by). We then issue a retraction.
Which read thusly:
G. Vianello (Stanford U.), D. Kocevski (NASA/Goddard), J. Racusin (NASA/Goddard), R. Ojha (NASA/GSFC/UMBC/CRESST) <[email protected]>, and S. Ciprini (ASDC) report on behalf of the Fermi-LAT team:
The analysis of new data downloaded from Fermi has shown that emission from the proposed high-energy counterpart of GRB 140330A (Vianello et al., GCN 16048; Pittori et al., GCN 16058) has continued up to at least 40 ks after the GBM trigger. This is very unusual for a faint burst such as GRB 140330A. Also, our localization of the LAT excess has improved, and it is now:
R.A., Dec. = 325.40, -64.13 with an error radius of 0.14 (90 % c.l., statistical only),
which is well outside the IPN error box (Hurley et al., GCN 16051).
Moreover, a search in Fermi archival data has shown a variable emission from a source compatible with our position, which was detected in several occasions in 2012 and 2013, and also detected on March 28 and 29, 2014. This source has been tentatively associated with the extra-galactic radio source PKS 2136-642 in the past (ATel #5695; Mauch et al. 2003, MNRAS, 342, 1117).
Therefore, we believe that the high-energy source detected by Fermi and by AGILE (Pittori et al., GCN 16058) is unrelated to GRB 140330A.
Further analysis is ongoing.
Vianello said the circulars are meant as trial balloons for precisely these sorts of observations. They aren’t peer-reviewed, and aren’t even papers:
Instead, they are a fast and easy way to exchange information on localization and characteristics of transient events (mostly GRBs), so that more instruments and observatories can be pointed and look while the transient is still ongoing. Thus, they are by definition preliminary, and all people using them know this very well.
Not sure this should be classified as a retraction. It’s a correction and these kinds should be welcomed and encouraged.
Omnologos, Vianello states “We then issue a retraction.”
A paper desn’t have to be in an academic journal, or to be peer reviewed, to be retracted. Any document that exists in the public sphere that is factually incorrect must be retracted, or corrected. I agree that this sounds like an internal open access report more than a traditional “academic paper”, but then why did Vianello retract it? Surely, a correction, followed by the explanation he provided RW, and a new report with correct information, would suffice?
I am rapidly starting to think, with all these retractions taking place, that scientific papers are becoming “transient events” and that we need more powerful telescopes to see them, further, and deeper in the black holes of the publishing universe.
Aren’t scientific papers always transient events? Just how long they last is variable. But, few stand the test of more than a few hundred years time, much less a few thousand years time. Eratosthenes did good work on measuring the size of the Earth; but it has been superceded by advances in measurement. Handwashing in Western medicine was sporadic until Semmelweis showed people a better way. Absolute frames of reference gave way to relative frames of reference with EInstein’s work. Notable exceptions are found in the history of Mathematics where a triangle on a flat surface behaves the same now as it did then ( See:
A History of Mathematics, Second Edition [Paperback] by Carl B. Boyer , Uta C. Merzbach, Isaac Asimov ).
So I, for one, embrace these changes as they are evidence of the actual change that is occuring.
Semmelweis was locked up in a nuthouse under the pretence of insanity while doctors who were unintentionally killing babies refused to heed his advice to wash their hands properly. He was beaten there, and died a few weeks later. He was a hero but his advice did not influence standard practice until after he became dead.
GCN Circulars are more like a mailing list. A search of NASA/ADS yields 43 entries with “retraction” in the subject out of 16,426.
Retracted because it was wrong. This is the right thing to do, because the signal was incompatible with a GRB. A correction would have implied that there was some sort of compatibility. Note that the data still stand. One of those rare occasions here where we should collectively raise our hats to the authors.
The GCN mailing list is not intended to be a final statement on high-energy astrophysical events; it’s a first draft of the observational record and the occasional correction is expected. The community of scientists that consume the GCN announcements are in agreement that rapid notices for these sorts of detections are important enough that a few false-alarms are tolerable.
Since the point of Retraction Watch is to disseminate information on what are generally duplicitous practices in the scientific community, I don’t believe that GCN announcements such as this should be included. Events of this sort are an example of how good science is done. At the very least, please make an explicit statement that the scientists who generate the GCN announcements should not be considered to be of the same class as the hucksters and charlatans who typically grace these pages.