Discussions about eggcorns and related topics
You are not logged in.
Registrations were closed for a long time because of forum spam, but I have re-opened them on a trial basis.
The forum administrator (chris dot waigl at gmail dot com) reserves the right to request users to plausibly demonstrate that they are real people with an interest in the topic of eggcorns. Otherwise they may be removed with no further justification. Likewise, accounts that have not been used for posting may be removed.
Thanks for your understanding.
Chris -- 2015-05-30
The post on shape of form began to steer in a new direction that seemed more appropriate for the Eggcornology section. I confess upfront that I’m venturing way past my pay grade here, to use a current catch phrase. Someone with the proper security clearance should show me to the door if I have pushed any buttons that should have stayed unpushed.
That being said, a discussion seems to be in order, to develop a convenient taxonomy for some of the species that show up when you go looking for eggcorns. In particular, in the discussion , a need was recognized for words for the reconstitution, reshaping, or transmutation of words and phrases at the hands of computer programs. The problem is that these potential “eggcorns” do not arise through any innocent human agency.
The suggestion was made in the shape of form post that a silicism would be a useful generic term for the following events:
Category 1 is now established as a cupertino. A series of posts on Language Log provide its provenance. I like it a lot. To provide a likely example, there are unaccountable but genuine reshapings on the web, which alter the standard blow the whistle to the unaccountable malapropism blow the weasel:
I write this not only to bark or ro to blow the weasel, but also as an opportunity to tell that big US attorney man that said that this is a nation of cowards that even thouhg I am scared to death to say all this, I am not a coward. (http://www.infowars.com/minority-report … cts-crime/)
They also bribed an employee of theirs for blowing the weasel on financial fraud at Microsoft (http://boycottnovell.com/2009/03/12/irc-log-11032009-3/)
This isn’t another gerbil thing, I hope! It might have arisen using Microsoft Word to check the spelling — though that seems unlikely for the first post — since, for example, if whistle were misspelled wesel then the first suggested “correction” is weasel in Word 2003. Hence, a cupertino.
I leave the floor, and the door, open for suggestions for categories 2 and 3. I’ve been struggling with these (see below in a conveniently avoidable annex). Arnold Zwicky has set out a remarkable example of how it’s done here . His self-confessed frenzy of naming generated, or if not, at least put in order, the terms pail, flounder, pineapple, floundapples, pailcorns, esculators, and Nunberg’s thinkos. The zwickian genius lies in the choice of juicy examples to represent a category, as was done for the eggcorn and cupertino. As an aside, kem has commented on pails recently. An example of the type 2 silicism is provided in the shape of form post, centred on silicism itself.
Wow, I see now that the Zwicky post was, in addition, a marvel of economy, and I will take a cue from that to split this post into “byte-size” pieces. To be continued.
Last edited by burred (2009-03-28 23:52:47)
I hope you do continue the post. It seems like it is going somewhere…
I can think of two more categories of silicisms:
(1) The odd phrases produced by automatic translations software. Pat has discussed a few of these in the thread at http://eggcorns.lascribe.net/forum/view … p?id=2440. We could call these Yalts, after the Alexander Yalt in the famous Monty Python sketch about the Hungarian phrasebook.
(2) When you do enough googling around the web to find odd words, you start to notice that the automatic reformatting of text leads to some real howlers. You get hyphens in crazy places, headers that become sentences, word lists strung together, etc.
Fusing a word that occurs at the end of a line with the word at the beginning of the next line is perhaps the most productive, if not the most common, of these formatting errors. Take the article at http://findinarticles.com/ao5eRROZn/402 … mnia.html. This article has been copied from another source. No one seems to have proofread the piece after pasting it into its new location. Among other errors, the lexicon of English words is expanded with these gems:
The first word in the list wouldn’t be bad type specimen for the species. We could call them acombinations.
Last edited by kem (2009-03-31 14:06:28)
Jorkel, I will show the jury that this thread is relevant.
Kem, Yalts and acombinations. Beauty.
This follow-up post on silicisms and eggcorns is organized as follows:
First, a story linking computers with eggcorns. Google, the search engine, was born of something eggcornish. The name for the startup, agreed upon by founders Larry Page and Sergey Brin*, was to have been Googol, a fanciful name for an astronomically-large number (10100). The story goes that it was “misspelled” on the registration form, as Google. Google, nevertheless, is a more fitting name for a search engine, for its closer connection to vision: google-eyed, goggle and go ogle (particularly apt). This was not a blind mistake. Pat would say it was eggcorneal. (See also http://eggcorns.lascribe.net/forum/view … hp?id=2766).
Second, misconstrual of the spoken word by voice recognition software, leading to eggcorn-like substitutions. EveryZing searches voice files for words and phrases you propose. Its current capabilities might be called “phenomenal.” If you enter a search for “misconstrue” in EveryZing, you get 40+ results, about 50% or so of which are badly misconstrued. For example, my search yielded misconstrue as a transcription of “it’s kind of strange” here: http://search.everyzing.com/viewMedia.j … =1&y=0&x=0
(click on the orange triangle along the blue timing bar to jump to the misconstrued snippet).
I discovered the EveryZing site while looking for “miasma theory” eggcorns. A large number of hits appeared for “my asthma” and “my asthma theory” from mildly or wildly misconstrued recordings that are searched by the voice-recognition program. These are not true eggcorn hits – they’re algorithmic screw-ups, but not unlike what humans might do… Google returns these Audio search engine results from EveryZing, based on “my asthma” or “my asthma theory”:
• Most to suppose to – my asthma theory which stated that disease was caused by foul – From rotting Manitoba inspect them want to.
• Same search as above: The result is a warm and racing swirling sonic my asthma with the center crystalline common uses it and all the songs – equally effective.
• Same search again: a kind of photographic evidence of you know of the sort of my asthma from beyond an act of – mistake could be.
All of these were translated from the spoken word miasma , which is clearly not in the program’s vocabulary, so that my asthma was the logical translation. Voice recognition must be extremely difficult for homonyms or homophones. That is my suggestion for a term to describe a voice-recognition software generated “eggcorn” hit – a myasthma. Many others are possible – EveryZing is a font of zingers.
Third, I couldn’t decide among many possibilities for a Google Books type scanner error. An example, as noted above, is proved by silicism itself. Here’re some ideas:
A zamt: a transcription error from antiquity. Latin at the time did not use a dot above the letter i. Transcription error of zamt produced the modern word zenith. (http://en.wikipedia.org/wiki/Spelling)
Next, a camel: erroneous interpretation by translators passing from Hebrew into Greek radically changed the nature of the biblical story about the chances that a cable might pass through the eye of a needle. (http://knowledgerush.com/kr/encyclopedia/Misspelling/)
Finally, from a more modern, computer reading error: a viagrant. “Computer misread prescribes Viagra instead of anti-smoking pill Zyban” (Headline from http://news.bbc.co.uk/2/hi/uk_news/scot … 175271.stm). The prescription for an anti-smoking pill Zyban, given by general practitioners, was misread by a computer, as Viagra. The name of the transcription system? GPass. The success of this zamt might depend on your answer to the question: Do you smoke after sex?
Next post: typosquatting and the fool sweep
Edited 6/04/2009 to include link to earlier extensive discussion on the nature of the word google
Last edited by burred (2009-04-06 22:34:53)
Jorkel, your silence with respect to viagrant was eloquent. Accepted. I’ve got another suggestion for OCR errors of scanned documents, a suggestion that might be familiar to sci-fi fans: a darkly. Philip K. Dick wrote a novel called, “A Scanner Darkly.” It’s a story of loss of identity and self-deception in a near-future dystopia. Nothing is clear and no one is who they seem to be at first view.
The title is one of many literary references to a verse of the Bible , from Corinthians, that says, “For now we see through a glass, darkly.”
The reference is to mirrors, not lenses, according to Wikipedia. Imperfect reflections through scanner glass.
Darkly is simple, stark, and conveys the idea of something clouded, obscured, poorly seen. These OCR errors are unique among silicisms in their capacity to misspell and distort words, to “invent” words not found in the dictionary. That capacity is in fact their essence, unlike myasthmas, cupertinos, and Yalts (acombinations are somewhere in-between) which are chosen from a word list. The inventions are not random. They are modifications of existing words. We may eventually find that there is a taxonomy of the darkly.
There is one important problem I see with darkly. That is that the plural, which was avoided above, is darklies, which is very unpleasantly reminiscent of “darkies.” Modification necessary. I will ask the machines. Are there words which scan as darkly? Or that yield darkly when asked? Duckly? Darkling?
The machines on Google Books had no problem with darkly – I’m ready to abandon it. There was an interesting transformation of “hardly” into bardly, however.
These two pages had double bardlies:
Also, from a book on Latin roots, I think.
“Stemmata Latinitatis” by Nicholas Salmon
nideo (bardly used) => I shine or glitter—I smile or smirk (originally from Horace)
(http://books.google.com/books?id=iFwSAA … zQSlpozsCg)
I apologize for turning these posts into a blog, it was never my intention. I will propose a solution and be done with it, and let the sheeps fall where they may.
So, I have made my decision. It is bardly.
Or darkle. Yes, it is bardly or darkle. Darkle, with back-story above, is a verb back-formed from the beautiful adjective darkling. The free dictionary says
v. dar·kled, dar·kling, dar·kles
1. To appear darkly or indistinctly.
a. To grow dark.
b. To become gloomy.
To make dark or indistinct: “the dramatist . . . whose province it is to darkle and obscure” London National Observer.
It is darkle.
Last edited by burred (2009-04-06 19:14:05)
1. A misconstrual of a word in a scanned document by optical character recognition (OCR) software. A type of silicism, notably one engendered by OCR.
2. (rare) By extension, the event of being misled by a darkle in the search for an eggcorn.
Etym. From 1 Corinthians 13:12 (King James) through Philip K. Dick, darkly, to adj. darkling > v. darkle > n. darkle.
No, the drug story is not clear. Perhaps there was a misreading or misprogramming of the pharma-code or barcode – looks suspiciously like human error somewhere along the line. Sorry, yes, the smoking comment was a feeble reference to an old joke – “I never looked”. The word viagrant was supposed to be a blend of Viagra, variant and vagrant. Confused in retrospect. Same for “zamt” and “camel” – not canonical examples, therefore ultimately they will not stand up.
(crossed responses) Oooh, I like Tuttle. That one can be sold.
The script for Brazil is here: http://www.dailyscript.com/scripts/brazil.html.
I like “buttle,” but it doesn’t quite fit the scanner imagery. A “buttle” would be more like a typo. More specifically, a typo that produces a sense word with interesting consequences.
Perhaps we are searching for a word that already exists. OCR errors first received widespread media attention in the early 1990s, when Apple introduced the Newton. The Newton was one of the first PDAs. Included with the Newton was a touch screen and handwriting recognition software called CalliGrapher. You could write on the Newton screen in cursive and the software would translate your scribbles into digital text.
Like most bleeding edge electronics, the Newton was pricey and buggy. The handwriting recognition feature was a massive underperformer. It worked so badly that Newtons became party talking pieces. An Apple Newton would be passed around and partygoers would write a phrase and ask the Newton to digitize it. The unpredictable results were sources of great merriment.
Gary Trudeau did a now-famous strip showing Michael Doonesbury using a Newton. You can view the strip here: http://images.ucomics.com/comics/db/1993/db930827.gif . In the last panel Doonesbury writes “Catching On?” and the Newton translates the question as “Egg freckles?” John Markoff refers to this strip in a 1995 piece for The New York Times, pointing out that the 1993 strip may have cost Apple millions of dollars and led to the early ouster of John Scully.
How about calling an OCR mistake an “egg freckle?” A small tribute to a happily defunct device and the cartoon strip that helped send it to an early grave.
Last edited by kem (2009-04-07 13:52:19)
You guys lose me with all those names. Maybe somebody can chart them out in some memorable fashion for the rest of us.
*If the human mind were simple enough for us to understand,
we would be too simple-minded to understand it* .
Darkles, Brazilian buttle-tuttles and egg freckles. Now I know what it is like through the looking glass. I don’t need to go ask Alice, however, I can ask jorkel and kem. I would let you two hash (or BC Bud) it out but I think the jury will decide for egg freckles. It’s even an OCR error. And the resonance with eggcorns is really too perfect. I suspect that buttletuttles will find a use somewhere further down
the rabbit hole the road.
I’m ready to submit my definition for fool sweeps later today and then I’ll make a table of these definitions so that we can all look them over.
(By the way, kem, thanks for the tip on foxy proxies)