Physics and Lexicography.

This announcement provides an example of science making a difference in the real world (which is to say, that of words):

QUT Senior Lecturer in Physics, Dr Stephen Hughes, sparked controversy over how a humble siphon worked when he noticed an incorrect definition in the prestigious Oxford English Dictionary.

In 2010, eagle-eyed Dr Hughes spotted the mistake, which went unnoticed for 99 years, which incorrectly described atmospheric pressure, rather than gravity, as the operating force in a siphon.

Dr Hughes demonstrated the science of siphons in a paper published yesterday in Nature Publishing Group journal Scientific Reports. […]

Dr Hughes, whose previous research has taken him to Bhutan to examine how siphoning could prevent inland tsunamis, said siphons had been used since ancient times but how they work was still debated.

“If you think of a car, atmospheric pressure is like the wheels, it enables it to work. But gravity is the engine,” he said.

“It is gravity that moves the fluid in a siphon, with the water in the longer downward arm pulling the water up the shorter arm.”

The Oxford English Dictionary corrected the error and removed the reference to atmospheric pressure after Dr Hughes pointed it out. However, he said the new entry “unfortunately remains ambiguous”.

“This definition still leaves the question open as to how a siphon actually works,” Dr Hughes said.

“But at least the reference to atmospheric pressure has been removed. The vast majority of dictionaries of all languages still incorrectly assert that siphons work through atmospheric pressure and not gravity.

Three cheers for scientists who pay attention to dictionaries, and for the lexicographers who listen to them! I have to point out, however, that the entry linked to is from an Oxford English dictionary, not the Oxford English Dictionary, whose entry is from 1911 and has a small-type section beginning “The way the action of the siphon is explained has varied” and citing explanations dating back to 1675.


  1. From the 2010 article:

    However, Ms Charlton said the 2005 edition of the Oxford Dictionary of English – a single volume dictionary and not to be confused with the much larger Oxford English Dictionary – did attribute a siphon’s operation to gravity.

  2. Stefan Holm says

    I share your three cheers for scientists as well as lexicographers – as long as our demands are within reason. Under an entry such as Light I would be satisfied with an explanation like for humans and most other animals: a sensation in the brain mediated through their eyes and caused by a limited spectre of electromagnetic radiation: a quantum mechanic phenomenon, the nature of which is not yet fully understood. Further discussions about the wave vs particle nature of light could be handed over to the physicists and the encyclopedias.

    Just like Stephen Hughes, who didn’t urge the OED to explain the nature of gravity (exchange of still undetected particles, ‘gravitons’ or ‘the curvation of four dimensional room-space’)? By the way – since air pressure also is an effect of gravity, it does have an (negligible though) impact on the siphon.

  3. I’m also annoyed by the constant misattributions to OED of the various activities of (especially) ODO and other ODs, but in this case it appears that OED is indeed the dictionary in question. The paragraph starting “The way the action of the siphon is explained has varied…” must be the correction described in the article (which explains Hughes’s complaint about the remaining ambiguity), since earlier versions do not have this explanation, and in fact described the siphon as being “used for drawing off liquids by means of atmospheric pressure”. This correction wouldn’t count as a full revision, which is why OED3 still says “This entry has not yet been fully updated (first published 1911).” Lots of minor revisions can have occurred to an entry between OED1 and OED3 (draft additions, e.g., and factual corrections) without a new publication date. There’s no real way to tell unless you have a paper or software copy of a previous edition.

  4. Ah, you may be right. Good catch.

  5. Such misattributions are inevitable, given that the OED is in part a loss-leader and marketing tool to sell smaller Oxford dictionaries to those who think There Is Only One Dictionary.

  6. David Marjanović says

    Rule of thumb: what dictionaries and even larger general reference works* say about technical terms is painfully wrong. Look up “dinosaur” anywhere this side of Wikipedia and weep.

    * I’m thinking of the German genre of Lexikon, sort of intermediate between a dictionary and an encyclopedia in the level of detail it provides.

  7. The case of scientific terms in historical dictionaries (like the OED) is especially interesting, since the historical dictionary wants to show what the term has meant at different times, and in chronological order. Scientific and technical terms, especially when overlapping with the general vocabulary (e.g. ‘light’), are like to have changed quite a bit over time. So that the first definition of (physical) ‘gravity’ in OED2, e.g., is “The quality of having weight, the tendency to downward motion, regarded in ancient physics as a property inherent in certain bodies (opposed to levity…)’ with quots from 1622 to 1678. This isn’t particularly valuable information about the physical force of gravity, but it’s still valuable information about the development of the concept, as evidenced by the usage of the word.
    Dinosaur, OED2 says, is among other things “something that has not adapted to changing circumstances; also, an object, institution, etc., that is extremely large and unwieldy” which is what some people say of OED2 itself. But that’s not quite fair, as evidenced by their patching up of siphon. Though dinosaur 1 remains the same as in 1896: “A member of an extinct race of Mesozoic Saurian reptiles (group Dinosauria, typical genus Dinosaurus), some of which were of gigantic size; the remains point to an organism resembling in some respects that of birds, in others that of mammals.”

  8. There’s no real way to tell unless you have a paper or software copy of a previous edition.

    There are “Publication history” and “Previous version” links in the online UI, which give you more than nothing.

    Not being able to see the details of online changes might well be a fundamental limitation, if they only scale access to the latest “published” out of their CMS.

  9. the remains point to an organism resembling in some respects that of birds, in others that of mammals

    Quite apart from the definition’s preserving of obsolete biology, what’s the function of “that of” in this sentence?

  10. @MMcM – yes, that’s right. For a detailed discussion of what can and can’t be ascertained from the online UI, see Brewer, “OED Online Re-launched: Distinguishing Old Scholarship from New” here:

    @Rodger C – it appears “that” refers to organism, i.e. an organism resembling the organism of birds or the organism of mammals. It’s a usage that strikes me as odd, too, but it seems to be out there in the Google books corpus (e.g. ” Origin of Uric Acid in the Organism of Birds”). I assume it is being used in a way akin to “skeleton” – i.e. the skeleton of dinosaurs, birds or mammals.

  11. marie-lucie says

    I think that the meaning of “organism” here is that of French un organisme, which is more widely used than the English word. It refers to the living body in terms of its internal organization regarding both form and function. The skeleton, the digestive system, etc would be parts of the same organisme (but not organismes themselves). The word is often used where English would use “body”: for instance, you could say that a given disease affecte tout l’organisme ‘affects the whole body’ (meaning all the internal organs). Le corps is ‘body’ but more specifically in terms of its external appearance: if an extensive skin disease affecte tout le corps, it is presumed to affect all or most of the skin, not the internal organs.

    Outside of science, the word is also used metaphorically for things like large, complex agencies set up for a specific purpose by a government or intenational ‘body’. For instance, UNESCO or UNICEF can be called organismes under the umbrella of the UN.

  12. Ah. I think “body plan” expresses more or less the same concept today.

  13. David Marjanović says

    “A member of an extinct race of Mesozoic Saurian reptiles (group Dinosauria, typical genus Dinosaurus), some of which were of gigantic size; the remains point to an organism resembling in some respects that of birds, in others that of mammals.”

    …Wow, the terminology used here is outdated in so many ways it’s actually fascinating:

    – “extinct” and “Mesozoic”; birds are dinosaurs.
    – “Race”.
    – “Sauria”. This name has largely been abandoned, and when it’s used, it’s usually used for the “lizards” (excluding snakes); when the name Dinosauria was coined in 1842, the dinosaurs were actually thought to be lizards, but hardly by 1896 anymore.
    – Uppercase for “Saurians”.
    – “Typical genus”; today this would be “type genus”, except that only family-group names have type genera. Even historically, Dinosauria is not named after any Dinosaurus.
    – “Dinosaurus”. Two different dinosaurs were named Dinosaurus after the name Dinosauria was coined; both times, this was invalid, because the name is preoccupied (by an animal that is not a dinosaur, as it happens). Additionally, at least one of the dinosaurs named Dinosaurus belongs to Plateosaurus, which was named earlier.
    – “organism” for “body plan” or thereabouts as explained above.

  14. birds are dinosaurs

    Technically, yes. But not even biologists take walks during the springtime to hear the dinosaurs chirping in the trees. The contract of a general-purpose dictionary like the OED is to define words as they are in fact used.

  15. marie-lucie says

    organisme = ‘body plan’

    Not quite. It is a body with a plan. If a disease, let’s say final stage cancer, affects tout l’organisme, it affects the entire body, not the body plan.

  16. The sense of organism in the OED definition above appears as sense 1 of the word, namely “Organic structure; = organicity n. Obs. rare.” (The OED1 definition is the same, except without the Obs.) Here are the quotations:

    1701 N. Grew Cosmol. Sacra ii. iii. §11 It is the advantagious Organism of the Eye, by which that is procured.

    1706 J. Evelyn Sylva 353 So astonishing and wonderful is the organism, parts and functions of plants and trees.

    1890 J. Martineau Seat Authority Relig. ii. ii. §3. 245 From the complexion of the language and the organism of the style.

    I was curious if this “J. Martineau” might be a French person writing in English, but no, James Martineau was the fourth-generation descendant of a Huguenot exile, and as English as could be otherwise.

  17. marie-lucie says

    Thanks JC. I checked organisme in the TLFI and my definition of the French word is correct (‘body composed of organs), but obviously the English word has (or had) a different definition (‘structure of such a body’).

  18. Not even biologists

    It seems I was wrong. Here’s a poem about it:

    Little did I realize
    That every summer breeze
    Brings the sounds of dinosaurs
    Singing in the trees.
    And in the cool of morning,
    When dew is barely dry,
    The cousins of Triceratops
    Soar across the sky.
    Triceratops is dead and gone,
    Which proves the worth of might.
    Maybe we should put our trust
    In music and in flight.
         —Eleanor Arnason (1974)

    Granted, Triceratops isn’t that closely related to avians, but it scans, whereas Tyrannosaurus doesn’t.

  19. David Marjanović says

    Wow. 1974 is really early!

  20. David Marjanović says

    It has often been said, and correctly, that birds are dinosaurs the same way that bats are mammals.

  21. marie-lucie says

    Davd: birds are dinosaurs the same way that bats are mammals.

    The large dinosaurs disappeared, and (some of ?) the smaller dinosaurs evolved into many species of birds. Even the largest birds such as moas and ostriches are much smaller than the large dinosaurs. Bats are still coexisting with other mammals, including quite large ones. Does it mean that if mammals (or most of them) became extinct, that would leave the field wide open for bats?

  22. David Marjanović says

    (some of ?)

    Only one of course.

    Even the largest birds such as moas and ostriches are much smaller than the large dinosaurs.

    They are not, however, smaller than the smallest nonavian dinosaurs…

    Does it mean that if mammals (or most of them) became extinct, that would leave the field wide open for bats?

    For those few bat species that are good at walking, notably these, perhaps – if the birds don’t beat them to it.

  23. From what I understand, the birds-are-dinosaurs thesis goes back to T.H. Huxley, but was thought to have been refuted in 1916 by Heilman, who thought dinos didn’t have fused clavicles (wishbones to the rest of us). With that factoid out of the way, Huxley’s views could be revived.

  24. David Marjanović says

    Heilmann published in Danish (Fuglenes Afstamning) in 1925, in the UK in 1926 and in the US in 1927, so you can find all 3 dates in the literature… 🙂 He thought they didn’t have clavicles (collarbones) at all. A furcula (wishbone) had been discovered in Oviraptor in 1924, but was misidentified for decades to come.

  25. John Cowan says

    I saw some New Zealand short-tailed bats on TV walking (waddling, actually) and burrowing the other night. Amazing. Apparently they are to ordinary bats as kiwis are to ordinary birds: they helped to fill the empty mammalian slot in pre-human New Zealand’s ecology.

    “Yawanna study ecology? Go learn differential equations.” —a professor talking to wannabe Greens back in the 1960s

  26. Wow. 1974 is really early!

    Not so much. Deinonychus (which was too heavy to fly) had a fused clavicle (the Mark of the Bird) that was discovered by J. H. Ostrom in 1969 already. Unfortunately, he buried the lede in this 172-page monograph, a detailed osteological study of Deinonychus. Ostrom was thought to be a little peculiar already because of his insistence on the endothermy of dinos, and that probably didn’t help, but “dem bones, dem bones gonna tell a tale”. Arnason was evidently plugged in to the palaeontologist gossip circuit.

    I thought of this today after seeing a TV program about moderately (Triassic and after) deep time, When Whales Walked, where someone was demonstrating how D. would scoop up prey in its too-short-to-run-on arms and crush them against its chest before dismembering and eating them, and showing us how this movement was in fact a flight stroke in advance.

  27. ktschwarz says

    The OED’s revision of dinosaur just went up. Here’s the new definition:

    A member of a diverse group of archosaurian reptiles (superorder Dinosauria), some of very large size, which dominated land habitats during the Jurassic and Cretaceous periods. Also (not in technical use): an animal of this kind, specifically characterized as very large and extinct; any of various other extinct reptiles of the late Palaeozoic and Mesozoic eras, esp. those of large size, including synapsids, plesiosaurs, ichthyosaurs, and pterosaurs.

    Dinosaurs originated during the Triassic period, and are known from fossilized bones, eggs, and other remains. They included quadrupedal and bipedal forms ranging in size from less than 50 cm to more than 20 metres, some having bony horns, crests, or body armour. Early studies compared them to lizards and crocodiles in mode of life, but they are now recognized to have had an elevated metabolism more like that of birds and mammals. In phylogenetic terms, the clade Dinosauria also includes birds, which are descended from feathered theropod dinosaurs.

    (The entry for bird, revised in December 2021, also mentions that “Birds evolved from a group of small theropod dinosaurs.”) Is that acceptable to a paleontologist?

  28. They’re letting synapsids into dinosaurs? “Not in technical use” indeed! And forsooth!

  29. I remind you that the OED is a record of general usage, not a science dictionary.

  30. David Marjanović says

    Deinonychus (which was too heavy to fly) had a fused clavicle (the Mark of the Bird) that was discovered by J. H. Ostrom in 1969 already.

    A clavicle is just a collarbone. In birds they fuse to each other, and the product is called wishbone or furcula. Such a thing was not recognized outside of recognized birds before the mid-90s.


    So long overdue (like… since the mid-70s to mid-80s at the latest) that it’s been obsolete for 20 years. We don’t use Linnean ranks at all anymore.

    Is that acceptable to a paleontologist?

    They should have made the “not in technical use” sense a separately listed sense. The second paragraph refers to dinosaurs in the technical sense only.

  31. They’re letting synapsids into dinosaurs?

    Dimetrodon has of course been a staple of cheapo “dinosaur” toy sets for the last ca. half a century at least. I’m not sure if this is about any other synapsids, though.

  32. David Marjanović says

    Dimetrodon is the only non-mammalian synapsid almost anyone knows, so that’s untestable.

  33. I did have a Paleontology Beyond the Dinosaurs book as a kid with a wider handful (including Edaphosaurus, Moschops, Lystrosaurus & Kuehneosaurus)… some of this sort also breaking outright into the ranks of “dinosaurs” in some source somewhere would be conceivable. Says right there in three of the four that it’s a saurus, right? 🙂

  34. David Marjanović says

    Kuehneosaurus isn’t a synapsid at all. It’s a diapsid, long thought to be a lepidosauromorph, but now coming out on the archosauromorph side. (It’s a bird! It’s a plane! It’s an allokotosaur!)

Speak Your Mind