ObscuripediaThe Free Obscure EncyclopediaThis article contains no fictitious entry. The question is reasonable and has been asked; the answer is that this encyclopedia has nothing that could be stolen without also being disbelieved, which is a form of protection the trade below would envy. Editors are asked not to add one as a joke, since a planted falsehood in an article about planted falsehoods would be indistinguishable from an ordinary mistake.
| What it is | A falsehood planted to be copied |
|---|---|
| Purpose | To identify plagiarists |
| Also called | Copyright trap, nihilartikel |
| On maps | A trap street; see Agloe |
| The name | |
| After | Lillian Virginia Mountweazel |
| Who was | A photographer of mailboxes |
| And | Not a person |
| Coined by | Henry Alford, 29 August 2005[1] |
| Does it work | |
| At detecting copying | Yes, demonstrably |
| At proving infringement | No |
| Because | Facts cannot be copyrighted[5] |
A mountweazel is a false entry inserted deliberately into a reference work – a dictionary, an encyclopedia, a directory, a map – so that any competitor who copies the work will copy the falsehood too, and can be shown to have done so.[1] The device is old and the trade is unembarrassed about it: an entry that is there to be stolen is cheaper than a lawyer. On maps the same thing is called a trap street, and the best-known of them, Agloe, stopped being fictitious. The same trap can be sprung without anybody setting it: the Mountains of Kong were copied from map to map for ninety-four years, and nobody had planted them.
The word is not a surname but a specimen. It comes from Lillian Virginia Mountweazel, who appears in the New Columbia Encyclopedia of 1975 as a fountain designer turned photographer, born at Bangs, Ohio in 1942 – a real place, which is the same care Agloe was printed with – celebrated for a collection of photographs of rural American mailboxes entitled Flags Up!, and dead at thirty-one in an explosion while on assignment for Combustibles magazine. She did not exist. Henry Alford took her name for the practice in The New Yorker in 2005, so that the technique is named after one of its own instances, which is Stigler's law of eponymy in its purest available form: the eponym not only failed to originate the practice but failed to exist.[2]
The best-documented modern case is a single word.
The New Oxford American Dictionary, from its first edition, carried esquivalience, n., "the willful avoidance of one's official responsibilities."[3] It was written by Christine Lindberg, one of the dictionary's editors, and it was given an etymology – a derivation from the French, invented along with the word. Remember that: it is the difference between this section and the next one. When Alford went looking through the letter E and asked, the editor-in-chief Erin McKean confirmed it: the word was planted, and had been there from the start, to protect the copyright of the electronic edition.
Two things about it are worth setting down. The first is that somebody had to invent a word meaning the wilful avoidance of one's official responsibilities and put it in a dictionary in order to find out whether anybody else was doing their job. The second is that it worked. Esquivalience duly appeared on Dictionary.com, credited to a different reference work, which is precisely and only what a trap is for.
Against that, the accident.
On 31 July 1931 the chemistry editor Austin M. Patterson sent Merriam-Webster a filing slip reading "D or d, cont./density", meaning to add density to the abbreviations taken by the letter D. Headwords were typed with spaces between the letters, so that "D o r d" read as one word rather than three. It was filed as a word.
Webster's New International Dictionary, second edition, accordingly carried, at page 771, between Dorcopsis and doré:
dord (dôrd), n. Physics & Chem. Density.
It was published in 1934. On 28 February 1939 an editor noticed the entry and became suspicious, and what he noticed was not the meaning but the absence of an etymology.[4] A plate change was ordered, density was properly added to the abbreviation, and the definition of the doré furnace was lengthened to close the space where the word had been.
Copies without it were bound from 1940 and copies with it went on being printed for years; it was not fully gone until 1947. Five years in an unabridged dictionary before anyone looked, and eight more to get it out.
The two cases sit badly together, and the discomfort is the point.
Esquivalience was put in on purpose by people who knew exactly what they were doing, and it was found because a journalist went looking for it. Dord was an accident that nobody planted and nobody wanted, and it survived five years in print. The accident lasted longer than the forgery, and it was caught by the same defect either would have had: no etymology, no evidence, nothing behind it. Nothing in the process had asked where the word came from. It had only asked whether it was already there, and it was.
That is a description of how a reference work actually maintains itself, and it is not confined to dictionaries.
A mountweazel establishes copying, and copying is not the offence.
The point was settled in Feist Publications v. Rural Telephone Service (1991). Rural, a telephone company, had salted its white pages with fictitious listings for exactly this purpose. Feist's directory duly reproduced 1,309 of Rural's 46,878 listings word for word, about one in thirty-six, and four of those were the traps. The copying was not in dispute and could not have been clearer.
Rural lost. The Supreme Court held that facts cannot be copyrighted and that an alphabetical list of subscribers has no original authorship in it to protect. The trap had worked perfectly. It had proved copying, which was never the question, and it had nothing to say about originality, which was.
The same holds on paper of a different kind. A trap street proves that a rival copied a map rather than surveying one, and cartography has had no better luck with it in court. So the device survives as an excellent detector and a poor instrument: it will tell a publisher, beyond argument, who has been copying, and then decline to be of any further use. It is kept up anyway, which is the most human thing about it. A reference work that wanted to prove what it had said, rather than who had taken it, would need a different device altogether: see the Incorrigible Register. A dataset that wanted to expose its own failures rather than a rival's would need Null Island, which is the same trick pointed inward.
00015ba17b5fecf868d6718e1f9e3783d42ce9514805bc35162c2fe44500ca2f