Oculudentavis: the plot thickens

July 24, 2020

Since we wrote about the putative tiny bird Oculudentavis (Xing et al. 2020) last time, things have become rather weirder. I want to discuss two things here: how we got to where we are, and what happens to the zoological name Oculudentavis khaungraae.

Xing et al. 2020, Extended Data Fig. 1. Close-up photographs of HPG-15-3. Part a, Entire skull in left lateral view. The black arrows indicate decay products from the soft tissue of the dorsal surface of the skull and the original position of skull, which drifted before the resin hardened. Scale bars, 2 mm.

First, how we got here. The timeline is a little confused but it seems to go like this:

  • 11 March: Xing et al. (2020) name Oculudentavis khaungraae, describing it as a bird. [link]
  • 11 March: In a Facebook thread on the day the paper is published, Tracey Ford claims that at least some of the authors were told at a symposium by lizard workers that their specimen was a lizard.
  • 12 March: Mickey Mortimer (very quick work!) publishes a blog-post titled “Oculudentavis is not a theropod”, making a solid argument. [link]; see also the followup post [link]
  • 13 March: Andrea Cau, working independently, publishes a blog post in Italian titled “Doubts about the dinosaurian (and avian) state of Oculudentavis” (translated), also making a solid case [link]
  • 13 March: Wang Wei et al. (the same authorship team as in the next entry) publish a detailed, technical Chinese-language article arguing that Oculudentavis is a squamate. [link] [Google translation]
  • 18 March: Li et al. (2020), in a BioRxiv preprint, formally dispute the identity of Oculudentavis, suggesting it is a squamate. [link].
  • 3 May: at the monthly meeting of the Southern California Paleontological Society, where Jingmai O’Connor gives the talk on “The evolution of dinosaurian flight and the rise of birds” she is allegedy “quite upfront about Oculudentavis being a lizard” [link]
  • 29 May: a note is added to the online version of Xing et al. 2020 stating “Editor’s Note: Readers are alerted that doubts have been expressed about the phylogenetic placement of the fossil described in this paper. We are investigating and appropriate editorial action will be taken once this matter is resolved.” [link]. (Steven Zhang later says on Facebook, “I’ve been reliably told by one of the coauthors of the Li et al. commentary piece, Nature rejected the comment from publication but then flagged up the matter as an Editor’s Note.”)
  • 14 June: O’Connor et al. (2020) (mostly the same authors as of the original description) reassert the avian identity of Oculudentavis. [link]
  • 22 July 2020: the original article (Xing et al. 2020) is retracted, with the reason given as “We, the authors, are retracting this Article to prevent inaccurate information from remaining in the literature. Although the description of Oculudentavis khaungraae remains accurate, a new unpublished specimen casts doubts upon our hypothesis regarding the phylogenetic position of HPG-15-3.” [link]

(Note: Facebook always seems very ephemeral, so here is a screenshot of the conversation in question:

I am aware that this is only hearsay, and rather vague: what symposium, what lizard workers? But I’ll leave it here as it does seem to be part of the story — judge it as you will.)

The unambiguously strange thing here is the O’Conner et al. preprint, published after O’Connor had seemingly accepted the squamate identity of Oculudentavis, but arguing for an avian identity. The O’Connor et al. rebuttal of Li et al. is pretty clear on its position, stating at the bottom of page 2:

Our parsimony-based phylogenetic analysis run using TNT placed Oculudentavis in Aves … Forcing a relationship with squamates required 10 additional steps.

But it also contains the rather extraordinary statement “Although in the future new information may prove we are incorrect in our original interpretation … this is in no way due to gross negligence” (p3).

I think we have to assume that O’Connor changed her mind between 11 March (the original publication) and 3 May (the SoCal meeting), then changed it back again by 14 June (the rebuttal of Li et al.), and finally accepted her first change of mind had been correct by 22 July (the retraction). But other interpretations are possible.

And of course the key question here lingers: why was the paper retracted, rather than merely corrected? And why does the journal say the authors retracted it, when the lead author says that the journal did it against their will?

Anyway, enough of the past. What of the future of the name Oculudentavis khaungraae?

The first thing we can all agree on is that (assuming Oculudentavis does turn out to be a squamate), the fact that the generic name misidentifies the phylogenetic position of the taxon is neither here nor there. Zoological nomenclature is full of such misnomers: they are not, and never have been, a reason to remove a name from the record.

But the retraction of the article in which the name was published is another matter. Does it mean, as some have argued, that the name is now nomenclaturally void?

I would strongly argue that no, it does not. There are several lines of reasoning.

First, the International Code of Zoological Nomenclature does not mention retractions at all — from which the simplest conclusion to draw is that it does not recognise them, and considers a paper once published to be published forever.

Second, the wording of the code pertains to the act of publication, not to ongoing status. In Article 8 (What constitutes published work), section 8.1 (Criteria to be met) says “A work must … be issued for the purpose of providing a public and permanent scientific record”. And the Oculudentavis paper certainly was issued for that purpose.

Third, the paper is still out there and always will be: even though electronic copies now bear the warning “This article was retracted on 22 July 2020”, there are thousands of copies of Nature 579 in libraries around the world. They can’t all be amended. What’s written is written. Quod scripsi, scripsi.

And this leads us to the final and most fundamental point: you can’t rewrite history: not one line. The simple and unavoidable reality is that the paper was published. That happened. A retraction can’t undo that — all it really amounts to is an expression of regret.

So the paper was published, and still is published, and the name established in it remains, and is forever tied to the type specimen HPG-15-3. If someone describes the “new unpublished specimen” referred to above, they have no choice but to use the established name Oculudentavis khaungraae: they don’t have the option of naming it (say) Oculudentosaurus instead.

At least, that’s how it seems to me. The International Commission on Zoological Nomenclature has been informally invited on Twitter to state a position, but has not responded at the time of writing — but then it’s not tweeted at all since April, so who knows what (if anything) is going on there? I heard somewhere that Oculudentavis is not being discussed on the ICZN mailing list, but I can’t remember where.

Now would be a good time for them to issue some guidance regarding retractions. And hey, ICZN? If you want to use any of my points above, feel free!



5 Responses to “Oculudentavis: the plot thickens”

  1. Pallas1773 Says:

    There’s more. On 11 March, Tracy L. Ford commented that the Oculudentavis authors had been told at a symposium that the skull belonged to a lizard: https://www.facebook.com/thomas.holtz/posts/10109039481184098?comment_id=10109039745838728&reply_comment_id=10109040399453878

    On 13 March, a Chinese article by Wei Wang et al. (same team who later wrote the Li et al. preprint) in a magazine made the case that Oculudentavis was a lepidosaur: https://mp.weixin.qq.com/s/e4j3E6bBxCDrjdxw2kXXQQ

    On 29 May, the Oculudentavis article got a note added, stating it was under editorial investigation: https://twitter.com/MarkScherz/status/1267476268662042624

    Steven Zhang commented that the editor note was added after the Li et al. response was rejected by Nature: https://www.facebook.com/thomas.holtz/posts/10109487000641518?comment_id=10109487308065438&reply_comment_id=10109488030073528

  2. Ian Corfe Says:

    One minor change to the timeline – the first version of the Li et al. 2020 bioRxiv preprint went up on 18th March (so submitted less than a week after publication of Xing et al since it takes a few days to clear and be published). The current, June 8th version is v.4 though I don’t think it’s that different to the original.

    I also thought I’d read somewhere that Nature reserves the right to serve a retraction if results are shown to be erroneous such that it invalidates the initial premise of the paper, which seemed a bit draconian, but I can’t find it on their site anyway so it may have been made up!

  3. Mike Taylor Says:

    Thanks, Ian, for that correction. I have updated the timeline in the article accordingly. (For anyone coming along after the event: the timeline previously had the Li et al. BioRxiv preprint coming out on 8th June, but that is in fact the date of the most recent revision of that preprint.

  4. Mike Taylor Says:

    And thanks, Pallas1773, for your additions to the timeline, too. I have updated accordingly.

  5. […] I’ve updated the timeline in Friday’s post to include several more events, kindly pointed out by commenters Pallas1773 and Ian Corfe. Check […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: