Phylogenetic roulette and the identification of sea monsters

Like most scientific communities, the world of academic tetrapod zoology is an incestuous place. Inspired by a comment made here at Tet Zoo by Matt Wedel - co-author, colleague and one of the three SV-POWsketeers - Cameron McCormick (who works on guppies but is collaborating with Michael Woodley and yours truly on sea monsters) has a new post up at Lord Geekington that I find pretty interesting. Employing the new term 'phylogenetic roulette', he laments the fact that people are inclined to latch on to highly specific identifications when mystery animals are reported, even though critical thinking and parsimony show that they've either gone way too far with the evidence they have, or have gone well astray in their interpretation, or both. As it happens, the doyen of sea monster research - statistician and fisheries biologist Charles Paxton - has been saying similar things in his lectures and published papers. As a case study, Cameron looks at one of the most famous sea-serpent cases of all time, the Daedalus encounter of 1848. Here is one of the famous Daedalus illustrations...


The Daedalus creature has been a 'classic' sea monster (viz, an unidentified giant marine animal representing a new, scientifically unrecognised creature) for some. As Cameron discusses, a specific yet very wrong identification was discussed by one prominent 19th century scientist. Others have proposed that the Daedalus creature was a giant squid. Naish (2001) - which I really must make a pdf of, if only I knew how to - says 'Most writers are familiar with the appearance of this creature due to the rendition published in the Illustrated London News under the supervision of the ship's captain. However, an original sketch drawn by Lt. Edgar Drummond reveals far less detail, and a creature far more distant, than the ILN illustration. Ellis contends, as did Lee in 1884, that the Daedalus serpent may actually have been a giant squid swimming backwards with the end of its tail above the water surface. Drummond's sketch does not discount this possibility and, furthermore, shows that our 'traditional' view of this account may be more colourful (and impressive) than it actually was. This sighting is thus not such a 'classic' after all.' (p. 76). Ah, dammit. We scientists and our pesky reliance on evidence.


On 'phylogenetic roulette', while the term is new, the concept has of course been around for a while: the late Clinton Keeling referred to it as grasping at straws. I don't think this is entirely fair, given that the people who propose unlikely identities for given cryptids typically latch on to what is, at the least, a very superficial resemblance. Naish (2001) - oh, have I mentioned this article? - includes a whole section titled 'A rash approach to identification' in which I lamented the fact that cryptozoologists have tended to go too far in proposing specific identifications for cryptids when 'it is not at all unscientific to conclude that an answer is elusive, or indeterminable based on the available data' (p. 79). Of course, this does not negate the fact that - in some cases - we can make broad-brush inferences about the sorts of animals we might be dealing with (Woodley et al. 2009), and remember too that there are at least a few cases where we can make specific identifications. The latter are exceptional, however (Paxton et al. 2004, Paxton & Holland 2005). Anyway, please see Cameron's article for further discussion.

For previous articles on aquatic cryptids see...

Refs - -

Naish, D. 2001. Sea serpents, seals and coelacanths: an attempt at a holistic approach to the identity of large aquatic cryptids. In Simmons, I. & Quin, M. (eds) Fortean Studies Volume 7. John Brown Publishing (London), pp. 75-94.

Paxton, C. & Holland, R. 2005. Was Steenstrup right? A new interpretation of the 16th century sea monk of the Ãresund. Steenstrupia 28, 39-47.

- ., Knatterud, E. & Hedley, S. L. 2004. Cetaceans, sex and sea serpents: an analysis of the Egede accounts of a "most dreadful monster" seen off the coast of Greenland in 1734. Archives of Natural History 32, 1-9.

Woodley, M. A., Naish, D. & Shanahan, H. P. 2009. How many extant pinniped species remain to be described? Historical Biology doi:10.1080/08912960902830210

More like this

By complete coincidence - honestly! - we looked yesterday at discovery rates among terrestrial mammals. All indications are that many species remain to be discovered. It should also be well known, and I hope it is, that the same is true for large marine vertebrates: recent discoveries, and…
On November 7th 2009, the Centre For Inquiry in London is hosting a one-day event titled Monsters From the Deep! It's being held at Conway Hall in Red Lion Square (a venue I know all too well...). I'm giving a talk at the event so wanted to advertise it: for more details please visit the CFI…
If you know anything about the literature on marine cryptids - or sea monsters, or sea serpents, or whatever - you will know of the Long-necked seal, a hypothetical mega-pinniped proposed by Bernard Heuvelmans (1968) as the explanation for sightings of giant long-necked sea (and lake) monsters.…
This weekend (17th-18th April 2010), the 9th European Symposium of Cryptozoology is being held at Engreux in the south of Belgium. I meant to attend and give a talk, but had to cancel for financial reasons. And it's just as well that I did, given that virtually all flights from out of the UK have…

To make a PDF--if the document is in Microsoft Word--do this:

OPen the doc. Go to file>print...; in the printer window there should be a button or dropdown menu that says PDF. Engage that button or dropdown menu, and it'll give you an option to 'save as PDF.'

Thanks. But I don't want to do that: I want to make a pdf of something that I will first have to scan. What I don't get (and what no-one explains) is how you 'stitch' the separate scanned pages together.

There is a program called "PDFmerge". you can stitch serveral 'loose' pages together. Comes in very handy, especially when dealing with scans.

Btw, this is my first comment on this wonderful blog. I check it daily and am always impressed by the superb quality of the posts.


In Adobe Acrobat (at least the version I use: 7.0), there's a "Create PDF" tab which then drops down a menu asking for the source of the PDF pages: if you choose scanner, you can then scan in each page separately.

When you say you want to stitch pages together, is your desire to stitch a double-page image in the book into a single page in Adobe? If this is your aim, I imagine Adobe Illustrator would be better suited for this purpose, since you can properly align and touch up images. You can then create a PDF from your newly stitched image.

My officemate (and fellow PhD candidate) recently interviewed Ben Radford (a lake monster expert) on his podcast, the Cosmic Tea Party (; admittedly, I haven't listened to it yet, though I'm sure it's good! I always found it funny when the hosts of dinosaur documentaries I watched as a child took legends like Mokele Mbembe seriously...

On topic... is it possible that the Daedalus "monster" was an oarfish? Or am I just being silly?

Mitch, doesn't that only work with Macs or can you do that operation with PCs now? I know that when I had to make a PDF of my thesis that I was provided with a very long list of instructions on how to make a PDF, which I completely disregarded as it was so simple with my old ibook.

The point about phylogenetic roulette seems obvious, but it apparently needs saying. We need measures of cryptozoologoical uncertainty!

Thanks, Simon and Steve. Steve: on the oarfish suggestion -sorry, but yes you are being silly :) In fact, sin of sins, identifications like this shouldn't be made, and you're guilty of playing phylogenetic roulette. The drawings and eyewitness reports of the Daedalus creature do not mention any oarfish-like features, so there's no reason at all to use oarfish as a potential identity, nor do oarfishes behave like the Daedalus creature at the surface.

Having a "save as pdf" option is not at all standard for Windows. However, there are freeware pdf-making and -merging programs for Windows out there.

By David MarjanoviÄ (not verified) on 21 May 2009 #permalink

If you don't have/want full scale Acrobat, simply scan your pages, save them as .jpgs and insert them sequentially into a word document. Then, as David suggests, help yourself to some freeware such as CutePDF, which installs as if it were a printer, so you simply select it in your Print function (it asks you where to save the .pdf file automatically). Print the document with all the images on it to e.g. CutePDF, and Bob's yer uncle.

"people are inclined to latch on to highly specific identifications when mystery animals are reported, even though critical thinking and parsimony show that they've either gone way too far with the evidence they have, or have gone well astray in their interpretation, or both."

- As I understand it, psychologists find that this is a very common pattern in human reasoning in general.

People pretty strongly prefer to defend their current ideas rather than discard them for new ones.

Hmm, apparently "confirmation bias" is the term I'm thinking of - .

Could giant squid hold part of the mantle above water? I thought they are very gelatinous animals.

PS. And yes, their eyes don't shine with electric light. ;)

Well, not a pdf then... How about a collection of your own articles on alternative subjects!?

By Laurence Crossen (not verified) on 21 May 2009 #permalink

Thanks for the shout-out on your blog Darren!

Jerzy: I suppose the gladius could keep the mantle fairly rigid in the event that it leaves the water. Even if we presume exaggeration in the reports, the portion of the Daedalus cryptid leaving the water, estimated to be ~3 m long at a height of ~1.2 m, sounds implausible even for a hypothetical Mesonychoteuthis with a 4 m mantle. The reported speed (12-15 mph), duration of the sighting (20 minutes), and the description of the movement by one officer as "steady and uniform, as if propelled by fins" does not seem compatible with the (probably weak) jet propulsion of giant cephalopods.

If you have Photoshop, it's easy to composite scanned pages. If you don't, send 'em to me and I'll stitch them.

Large squid come in two kinds, the blobby kind, such as Mesonychoteuthis, and the streamlined kind such as Architeuthis and Taningia. Colossal squid, of course, would not be able to hold themselves above water like that, but the other kind might be able to; certainly Taningia danae seems quite rigid and a powerful swimmer from the videos I've seen (which are absolutely amazing, by the way).

The problem, of course, is that all these large squid are deep-dwellers, only rarely coming to the surface and generally because they've been caught by something, usually the ship they're being viewed from.

I think part of the problem with such sightings is the way seeing works. The brain doesn't just dispassionately assess the actual visual data supplied by the eyes (which are often markedly less detailed than one imagines); it manipulates and reinterprets them quite extensively in terms of both timings and geometry according to what, on past evolutionary and personal experience, "makes sense". Consequently when someone sees something novel, their brain is liable to present it to their mind in more familiar terms, so they may honestly think they saw characteristics and features that were not genuinely present.

This results in various manifestations of pareidolia, from Rorschach interpretations to misperceiving objects - especially in low light and/or when tired, and often as people. A personal anecdote by way of illustration: when driving suburban streets late at night, I sometimes think I see one or more persons standing at the roadside, only to realise that it's a pillar box or other piece of street furniture. The oddest thing is that I don't just perceive 'a person', but 'a middle-aged woman with a slight stoop, wearing a thin beige windcheater, a knee-length dark skirt and a worried expression' or some equally detailed image.

While I am, of course, mad, I don't think this is very unsual, and there are (I assume) sound evolutionary reasons for the brains of humans (and any other potential prey species) to behave like this - repeatedly misperceive patterns of leaf shadows as leopards and you'll just waste a little extra energy; misperceive just one leopard as a pattern of shadows, and you're liable to become leopard lunch.

A further difficulty is that memory is not a fixed record but a dynamic process, so when we "recall" things our brain is liable to edit the recollection according to what makes (more) sense to us. It's also very susceptible to being influenced by what others say. This is why witness testimonies (in court or elsewhere) are often both honest and incorrect, and why witness statements (whether of a crime or a sea-serpent) should be recorded as soon as possible after the event.

By Terry Hunt (not verified) on 25 May 2009 #permalink

Hey Terry,

what a nice summary of Heuvelman's :The Metamorphosis of Unknown Animals into Fabulous Beasts and of Fabulous Beasts into Known Animals, Cryptozoology: Interdisciplinary Journal of the International Society of Cryptozoology 9 (1990), 1-12!

MistarX @ 17: Really? Sounds like something I'll have to get hold of: I'd say "great minds think alike," except that clearly his was and mine isn't. Just to dispel any doubt, the above was an entirely off-the-cuff synthesis of factoids I've picked up (mostly from New Scientist, probably). I think the riff on leopards is original - though consciously inspired by someone or other's claim that leopard predation has been the numerically largest cause of death in hominid history - but I've read so much over the last 47 years that almost any idea surfacing in my thoughts might actually be someone else's, read long ago and forgotten.

By Terry Hunt (not verified) on 26 May 2009 #permalink

I remember reading in a recent book that the most likely explanation for this particular sighting would be a leopard seal,especially given the location of the sighting. Actually I think this makes more sense than a squid swimming with its tail out of the water.