Back in 2011 I posted about “an enjoyable and useful vocabulary test that gives you a bunch of words, asks you to check whether you know them, and extrapolates your total vocabulary size” that attracted quite a bit of interest (results here); now Ghent University has put online a similar test, introducing it thus:
In this test you get 100 letter sequences, some of which are existing English words (American spelling) and some of which are made-up nonwords. Indicate for each letter sequence whether it is a word you know or not by pressing the F or J key. […] The test takes about 4 minutes and you can repeat the test as often as you want (you will get new letter sequences each time).
If you take part, you consent to your data being used for scientific analysis of word knowledge.
Advice! Do not say yes to words you do not know, because yes-responses to nonwords are penalized heavily!
Advice! The test works best in Firefox, Chrome or Safari.
My results:
You said yes to 83% of the existing words.
You said yes to 0% of the nonwords.
This gives you a corrected score of 83% – 0% = 83%.
You are at the top level!
Theoretically, I could have done better, since I said “no” to a number of items that I thought could very well be words and turned out to be, but I didn’t want to risk yes-responses to nonwords (“penalized heavily!”), and the words were so obscure (“a rare name for the hyrax”) that I don’t feel bad about not knowing them. Enjoy! (A tip of the Languagehat hat to Trevor Joyce, who sent me the link — and who got 96% with no false positives, the bum.)
By contrast I threw caution to the winds and got 96% (all but three) of their supposed actual words but gave false positives to 17% (five items) of their supposed non-words. One of the actual ones I missed looked to me instinctively like a wrong spelling of a real word, but in hindsight maybe I should have assumed that they weren’t testing whether you knew which of X-able v. X-ible was the “real word” for a given X? The false positives all look to me like perfectly cromulent words and at least one of them looked like it might be a rare synonym for hydrax, which was something I had somehow been primed to be on the lookout for.
I note FWIW that the lexicon of English is so vast that no native speaker, however well-educated and/or idiosyncratic (let’s say a serious scrabble player, for example), will know it all. So “X is definitely not an English word” is not actually a sort of knowledge native speakers will actually reliably or consistently have. Now, maybe their differential results on their fake words will provide useful insight into which seem more plausible (i.e. fit expected orthographic and phototactic patterns smoothly) than others, and maybe that’s one of the things they’re actually trying to measure?
There are I suppose some actually-existing English words out there that maybe seem like they shouldn’t be because they are exceptions to otherwise strong patterns of what is and isn’t orthographically/phonotactically plausible. I’m not sure if they’re testing for those – nothing like that leapt out at me from the assortment I was offered.
96%, forewarned by Hat about the weighting: I okayed no non-words, but timidly rejected the perfectly cromulent “toluidine”, “disconcertment” and “kaikatea.”
On reflection, I should have adopted the view on “disconcertment” that, if it wasn’t already in the dictionary, I was adopting the prerogative of a native speaker and thereby creating it.
But Maori trees exist (in English) only for the purposes of cryptic crosswords. There, je ne regrette rien.
It was 69% real words and for 0% non-words for me. I was so careful and scared by the heavy penalties that I rejected the very first word, which happened to be “matchstick”. Not sure what I was thinking. Over time the pattern became clearer with the non-words being much more obviously bizarre, rather than compositions which could easily be words but accidentally noone coined them yet.
Hm, 94%. No non-words clicked as words. Notable miss was “impendency,” which is of course a perfectly cromulent word.
The nonwords are interesting; a lot of them were built entirely plausibly. Like others I was being cautious. English moves so fast.
With less timidity I got 90% on a second attempt, with 93% correctly identified real words and 3% from one accepted fake, which was “menincaine”. That would be a pretty neat name for a drug, I am a bit surprised it hasn’t been taken yet.
Also they sort of exaggerate about the “heavy penalisation”, when pretending to know a nonword costs exactly as much as not knowing a real word.
I was timid, so no false positives but wrongly rejected “donnish”, “convictive”, and “chauffer”. I got 90%. One of the words I did get was “illative”.
94%, with no false positives but missing 4 of the real words, to wit:
windsucker — a horse with the ‘injurious habit’ of swallowing air
comptometer — pre-WW2 word for a calculator or computer
lierne — an element of gothic architecture
demulcent — soothing
These all looked plausible but as ES said so did a lot of the non-words.
96% of the real, 0% of the imaginary. I simply pressed yes when I “knew” a word, and no otherwise – per instructions. A few plausible “sequences of letters” tempted me to say yes, but I hewed (?) to the truth.
Actually I’m a nervous wreck now.
I may have just lucked out – there was no nonsense about “lierne” and “demulcent”.
Surely it can’t be the case that honesty is the best policy ? Maybe only with vocabulary tests.
99%
99% + 0%
Good guesswork, I suppose.
I got 94% true positive, 0% false positive. There were two more words that I wasn’t certain about, so I wrongly rejected them, including hygroscopic, which I should have gotten. There was actually one non-word that was only one letter away from a real word, which seemed like something they ought to have avoided. On the other hand there was also one word that I thought looked phonotactically impossible for English (to the extent that anything can be phonotactically impossible in English, with its vast hoard of loanwords).
And I tried it again and did a little better, only missing two (97%).* What was interesting the second time was that both words I missed I could have gotten. They were uncommon adjectival forms based on nouns that I did know: gneissose and beechen.
* So a false positive evidently deducts twice as much from your score (give or take) as a false negative (which is the correct weighting).
“I could have done better, since I said “no” to a number of items that I thought could very well be words and turned out to be,”
I suspect that this is the intention. It seems to me that they are trying to determine people’s usable vocabulary (active or passive), not their pattern recognition skills.
Keep in mind that we are contributing to a database, not trying to gain admission to the university of our choice.
rather than compositions which could easily be words but accidentally noone coined them yet
I actually got one of those: oresquads. Also releaux, IIRC, which sounded realistic-ish, but (again IIRC) all the others were very obviously fake.
I rejected about half a dozen real words, but forgot to write down which ones (the only one I still recall is beaux). My score was 89%/0%.
Their fake words generator is kinda obvious. I got 0 false words, 87% real ones, but I totally guessed about a third of them using a simple heuristic: if I can quickly come up with a plausible pronunciation then the word is probably real even if I don’t know it
The one real word that I missed was chronopher, which turns out to be “An instrument that transmits the correct time to distant points by means of electricity.” This definition had a strong whiff of the 19th century about it, so I did some research. An ngram shows that it was in use from 1860 to 1920 or so. It appears that there may have been only one chronopher, and it telegraphed the time of day from the official timekeeping service in Greenwich to points in the UK (and perhaps further away) like post offices and railway stations.
As for words that are obviously fake: that’s very hard to tell. Is bordereaux an English word? How about omeprazole? Is bidon? is scree?
OK yeah this is 100% a psycholing experiment that has to do with response times; everything in the setup screams it (e.g. the penalty is intended to make you think about the tough sequences). There’s no actual attempt to score people on a comparable scale – but it’s well-hidden and the result is fun!
if I can quickly come up with a plausible pronunciation then the word is probably real even if I don’t know it
That would surely have been misled by both releaux* and oresquads.
*) I’m not actually very confident of the exact spelling I got, but I do recall it sounded plausible enough; when I got the (real) word beaux later on in the test I was similarly suspicious for similar reasons
A note on one of the words above:
A comptometer is a kind of adding machine with a very wide carriage for entering very large numbers, and it was around for quite some time after WWII, killed off only when computers came into common use. My first job was working as a secretary at Doubleday & Co in 1962. The first week I was there, the research dept held a goodbye party for Lucy, who had been a comptometer operator there for 50 years. I remember it vividly because I was only 16, and the idea of walking in the same office door for 50 years scared me to death!
Oh, and 93% for me.
Interestingly, I also got 83% of the real words and 0% of the nonwords.
The timing was the interesting part. It took me especially long to reject “futlord” for some reason.
Once I did another one of these computer tests, a non-linguistic one. It used reaction time to estimate your age. It got mine exactly to the year, which was sobering.
Futlord is obviously youthspeak for what we Old People call a “lothario.”
Pooded is clearly a perfectly good word, too, despite the fact that this test anathematises it.
“The many-pooded produce of the fertile farmlands of our glorious Motherland” (as we say in the UK.)
I thought it was a futbol champion.
The meanings naturally overlap to some extent.
I took two more passes. On the first I managed to get 100% for recognizing actual words, but at the cost of a 20% rate of false positives, so my approach needed a bit more tweaking. On the second I got my false positives down to 0%, with a 96% on actual words (meaning three false negatives). My false negatives were “abdicant,” “acatalectic,” and “flexure,” all of which I would have guessed were legit if I hadn’t been trying to avoid such guesswork. The reaction-time data is interesting. On the last try the two actual words I took the most time before giving a thumbs-up to were “tortuosity” and “intervallic.” The nonwords I took the most time before rejecting were “acelent” and “clocover.”
A comptometer is a kind of adding machine with a very wide carriage for entering very large numbers
I know the machine but didn’t know the name. When I was in grad school there were a couple of these devices in one of the common rooms. My thesis adviser had used them to calculate stellar models. They had a variety of buttons and settings, with a handle to turn the mechanism, like the spindle on a mangle or a butter churn. We devised a silly game with them. We would set them to zero and then, with a mighty push on the spindle, see how high a number we could spin up. I forget what record was set by the time I left.
87-0.
Like JWB with disconcernment, there were a certain number of randomly anglo-suffixed words that I felt distaste for. I gritted my teeth and yessed them till balking at dinnerless for no reason other than it’s ugly. IPhone Spellcheck hates it too. Not a big fan of cartelist either.
Should have been more confident on hendiadys, my longest pause at 4.2 seconds, but I was wary of the negative penalty. And followed my instinct that the parts of fibroblastic added up.
When I saw “gobstopper”, it was like seeing an old friend. I only know the word from Roald Dahl, but he describes the thing so vividly.
(I think they are what Americans call jawbreakers, only maybe even bigger.)
@Y: I’m not sure I could recognize a gobstopper in the wild, but my 6-year-old is a big Roald Dahl enthusiast so anything that is in Dahl’s lexicon definitely qualifies as a legitimate English word in our house. (His oldest sister, now 19, was a big Roald Dahl enthusiast in her day, but at a somewhat older age when she could just read the books entirely on her own, and thus had less of an impact on the idiolects of the rest of the household.)
Do you have a copy of Boy?
96%. They said nasogenesis is not a word, but:
http://www.amritas.com/080712.htm
Xerîb: Good thing they didn’t use dimidiation (from your link). I’d have missed it for sure. Or even worse, nasoexodus.
@J.W. Brewer: For the somewhat older Roald Dahl enthusiast (nineteen is probably a good age), there is “Lamb to the Slaughter.”
Fifteen years ago, we were over at my wife’s then closest friend’s house, and we discovered that both of the women were huge fans of “Lamb to the Slaughter.” However, neither my wife nor I had seen either of the teleplays of it. So at our hosts’ suggestion, we watched both of them back to back.
Lamb to the Slaughter is the only story of his I can think of which seems to be set in America.
Oh dear. I did worse than any of those of you who have told us what they got: 60%. However, the false positive that I got (“manicessly”, or something like that) was due to pressing the wrong key by accident. I knew perfectly well it was a non-word.
79% with no fakes checked as real. A bit flummoxed with their IDing “multi” as a word. Go ahead use “multi’ in a sentence! But, on the other hand, I appreciate finally knowing “eth.”
I did choose to display all on FB, noting to lure in my friends, “Preparing best I could for either a humbling experience or a bit of ego-polishing, just now (4:47 am), I took this test that scores me on the % of English words I know.”
Thanks for the post. A great opportunity to learn new words I can use in Scrabble to beat my better 2/3.
@David E “Kaikatea”
Are you sure that’s the spelling Ghent gave? Because it’s wrong. The NZ ‘white pine’ is “kahikatea”. Definitely 5 syllables.
But Maori trees exist (in English) only for the purposes of cryptic crosswords.
Em no, they are also mighty giants of the forest (and make beautiful furniture on which to display my Grandmother’s heirloom dinner service).
I may indeed have mistranscribed the original.
I must also concede that such trees are good for heirloom displays as well as crosswords, though I have no first-hand experience of this usage.
I got 78 %, although I clicked yes on some words only because they sounded vaguely scientific, not that I know what they mean. 81 % yes to existing words, 3 % yes to non-words (one single word, it turns out). It took me a while to decide on the nonsense words, many of them seem like perfectly normal words with unusual spelling. And I wish ullteemed was a real word! But what would it mean?
“To ullteem” means to enhance in some manner the last element in a sequence. You are surely familiar with ulteemate stress ? “Ullteem” has it, when used as a verb.
The adjectival form marks a final exercise of maternal patience: “I’ve told you for the ullteemth time to clean up your room!”
Nasogenesis is the theory that the universe was sneezed into existence.
I got 97%, which I suppose is reasonably okay for a native English speaker. I made a fat-finger error in rejecting dumdum, which I know very well to be a word. I agree that the real test is obviously of response time.
81-0, using the same approach as LH with similar results. The only real-word I rejected which was not a plausible compound was “ramie”. Most of the non-words were obvious. Perhaps my longest rejection time was for ch_che (with an underscore).
Nasogenesis is the theory that the universe was sneezed into existence.
When I was a kid, that is how I was told the primordial gods Shu and Tefnut were created–Atum sneezed out Shu and spat out Tefnut. This is apparently a later formulation. Utterance 527 of the Pyramid Texts says: “Atum created by his masturbation in Heliopolis. He put his phallus in his fist, to excite desire thereby. The twins were born, Shu and Tefnut.” And there are yet other scenarios.
Good thing they didn’t use dimidiation (from your link). I’d have missed it for sure. Or even worse, nasoexodus.
Although I didn’t know nasogenesis and nasoexodus, for some reason dimidiation is familiar to me.
I took it twice, and the first time, I got three that I hit “yes” that they were words that weren’t, but, looking at those non-words afterwards, all 3 were me hitting the wrong button by mistake. (On my phone, so it was tapping a button on my screen, rather than pressing a key on the keyboard.) The second time, only one, and this one I was either another case of hitting the wrong thing, or else I misread the word.
I did, taking it, assume that the non-words weren’t simply spelling errors, though I chose “no” for waling anyway, since I didn’t actually recognize it (and, yes, it’s a word).
In some cases, it’s not so clear cut whether one “knows” a word. Macrocosm was one. I can’t recall having encountered it before, but it’s rough meaning is clear enough.
Taking it one more time just now, I used the strategy of “yes” if the word conjured a meaning in my head, and I managed to make no yes-that-should-be-no mistakes, and got 86% of the real words.
By coincidence I just came upon what might be a good word for their test, in that it certainly seems like it could be a cromulent technical term in some scientific field you are unfamiliar with, but “could be” !+ “is.” And indeed googling shows that “dilatation” has been in some degree of use over the last century in certain medical and engineering contexts rather farflung from the context in which I spotted it, which was the statement that “there is the land of spices and perfumes, the dilatation of God’s goodness.” (From a sermon by John Donne, preached at Whitehall on March 4, 1624.)
This maybe wasn’t a good example but it made me wonder if there are strings-of-letters out there that the Ghent people deliberately avoid using in either of their categories, just to sidestep is-it-really-a-word controversies. Archaic/obsolete lexemes might qualify. On the other hand, some of the is-a-word items they did use the times I did the test also seem susceptible of controversy, i.e. proper names or brand names (but spelled lower case) or not-very-domesticated loanwords. I took sort of a meta-guess about test design and theorized that e.g. they’re not going to put in fake loanwords, e.g. a word that looks obviously non-English but also looks like a standard romanized spelling of what might or might not be an actual Japanese word but is perfectly cromulent from a Japanese phonotactics perspective whether or not it’s actual. I think the same test had a lowercased string of letters than when capitalized is an actual toponym of a municipality within an hour’s drive of my house and my instinct likewise was “I don’t know whether or not that’s an English word in any sense other than that toponym but I don’t think they’re going to treat it as a nonword given the toponym” and indeed they did treat it as a real word. Beyond the toponym I now learn that it also apparently refers to “an extensive flowering plant genus in the spurge family,” but of course “spurge” totally feels like a fake word.
In the 2019 national spelling bee, there was an 8-way tie. These were the final words:
auslaut; erysipelas; bougainvillea; aiguillette; pendeloque; palama; cernuous; odylic
If they had shown up in the Ghent test, I would have said that #s 2, 3 & 4 are words, and I would have nixed the rest.
Surprisingly, nasoexodus is not a medical term. It’s primarily used to describe the rejection of Ovid’s poetry in late classical times – Wiki.
@Bloix: But in the spelling bee context they give you obscure words but not fake words. So one of the key skills at the margin, beyond having memorized an absurdly large number of words and their spellings, is the ability to accurately guess at the spelling of an obscure word you don’t actually know from hearing it pronounced (and pronounced is sort of a stylized/over-articulated sort of way that probably makes guessing at spelling easier than if you just overheard it in someone else’s conversation).
@JWB:
“Dilatation” is common in ophthalmology, as indeed is the practice itself.
(In fact, using “dilation” marks you as Not One of Us, just as surely as using the spelling “opthalmology.” It’s one of the first things they teach you …)
There are a few other Dahl stories set in America (The Way Up to Heaven and Pig, at least).
Fun times.
Hemmed and hawed over ‘staggerbush’ for a long time. It looked extremely plausible, but I eventually admitted I didn’t know it. Dammit, it was legit.
Hysterokinetic would be a great word with which to trip the unaware.
@David E.: But there’s no backformed verb? A few years ago I spent considerable time at the offices of various specialist eye doctors (one for retinal-tear issues, the other for cataracts, since I was having multiple difficulties at once), and don’t recall ever being told by the staff after being given drops that the doctor would see me in 15 minutes after my pupils were fully dilatated.
They were speaking to you in Optometrist Trade Jargon.
OED s.v. dilation:
It’s a shame we’ve given up on dilation “Delay, procrastination, postponement,” last seen in the 17th c.:
I like that Halliteration: “desperat debters … dilation … disability.”
Dilatory is still around, in fancy language.
Yes, I use it myself when I want to sound fancy.
In scalar-tensor gravity theories, the extra particle/field is the dilaton, never the “dilataton.” Some people do use the etymological “dilatation” instead of “dilation” for the corresponding scale-changing coordinate transformation though. I advised a native German speaker colleague to use “dilation” in a paper, but he didn’t listen.
J.W. Brewer:
Everything you say is true. My point, poorly expressed I suppose, is that it’s trivially easy to compose a list of words that even the most well-read of us cannot identify as genuine. Pendeloque? Seriously? Odylic? Geddoudaheah.
Odylic.
I rest my case.
I scored 90%: 93% gross less 3% for “fluxness.” I did my best to avoid marking non-words as real but “fluxness” was begging for recognition.
Victor Mair.
fluxness
I’m sure that it must figure somewhere in the work of
https://en.m.wikipedia.org/wiki/Oliver_Heaviside
(see under “Electromagnetic terms”)
Victor Mair
His manner can indeed sometimes be a bit dimidiating.
I think it is one of those meta-experiments to observe what the participants say about the experiment afterwards, not what they do in the experiment. This thread is being monitored in Ghent.
Like that tv commercial set in an Italian village, where 100 bedsheets are trampled by bulls goaded by 10 lusty young men, then washed in Dash and hung up blindingly white in the streets. The selling point is the final interviews where the old women gush about the results. You could have had these without the spectacle, but it wouldn’t have been anywhere near as convincing.
How They Brought the Good News from Languagehat to Ghent.
dimidiation … Vicor Mair
Hey, as anyone who is able to read the results of a google search knows, dimidiation is superseded by impalement (and further search shows that it is an alternative to quartering). Be grateful for what you have.
Terminology involving flux generally goes back to Newton. He called his formulation of the calculus* the “method of fluxions.”
* The name calculus itself came from Leibnitz’s first (1684) paper on the subject in Acta Eruditorum, “Nova methodus pro maximis et minimis, itemque tangentibus, quae nec fractas nec irrationales quantitates moratur, et singulare pro illis calculi genus” (“A new method for maxima and minima, and for tangents, which is not impeded by fractional or irrational quantities, and a singular sort of calculus for this”).
I don’t know how many scientific traditions have specialized ML calculus to the infinitesimal calculus. Kalkyle in Danish (normatively kalkule) is any sort of calculation (mainly of cost) and in high school differentiation and integration were taught separately, while at university it was all lumped with series and some topology and called analyse. Though you can talk about infinitesimalregning if you insist.
Oresquad: a squat oread. Kaikatea: the opposite of ataraxia.
It informed me that I only marked one nonword as a word, namely, ‘seconds’. However, I wonder how I would tell someone I spent 47 seconds taking this test if ‘seconds’ is not a word?
@Lars Mathiesen: In English, calculus still has the general meaning of “a method or system of calculation.” The OED has not updated its entry, and it treats, “the calculus,” the specific differential and integral calculus due to Newton and Leibnitz as merely an example of this general meaning. The OED also lacks the newest meaning of the word completely: “decision-making method, especially one appropriate for a specialised realm” (per Wiktionary); I think this is the second-most common calculation-related usage (appearing particularly commonly in the phrase “political calculus”) of the word,* after the mathematical analysis sense. (Analysis is used in English to denote the larger mathematical field of which the calculus is a part. Analysis, dealing with continuous behavior, is one of the four principal areas of pure mathematics, alongside logic, algebra, and topology.)
The OED does have some interesting seventeenth- to nineteenth-century citations for calculus though, such as this one from Thomas Carlyle: “Science, which cannot, with all its calculuses, differential, integral, and of variations, calculate the Problem of Three gravitating Bodies”; or this rather mystifying one from George Eliot: “Fount of spirit force Beyond the calculus.”
* There is, of course, a whole separate group of mineralogical and dental meanings related to mineral accumulation.
Do practicing mathematicians nowadays use anything other than “calculus” in general, plus “the calculus of variations”?
@Y
https://en.m.wikipedia.org/wiki/Functional_calculus
Archaeologists love dental calculus. These authors were looking for evidence of diet and found a surprise:
Medieval women’s early involvement in manuscript production suggested by lapis lazuli identification in dental calculus
Kalkül and kalkulieren have only metaphorical meanings; I was taught to differenzieren and then integrieren.
a whole separate group of mineralogical and dental meanings
And renal. Ouch.
Kalkül and kalkulieren have only metaphorical meanings
Yes for the former, but not for the latter – it can mean to calculate, especially in economic matters (e.g. a Kostenvoranschlag “cost estimate”), or prices and tariffs. In that case, the verbal noun is Kalkulation.
Uh, yeah, that was very imprecise of me. Soporific weather lately.
I was thinking of einkalkulieren “take into account”.
Kalkül is apparently one of the German technical terms corresponding to formal system, and indeed many formal systems are called the/an X calculus in English, the most familiar to me being the relational calculus (although I much prefer the equivalent relational algebra, which is much easier for me to understand).
…Yes, says Wikipedia, and in that sense it’s suddenly masculine instead of neuter. I had no idea.
Similarly, die Erkenntnis is the subject of epistemology, das Erkenntnis is the verdict of the Constitution Court.
das Erkenntnis is the verdict of the Constitution Court
That’s only so in Austria. Duden says the neuter with the meaning “court verdict” is “Austrian, otherwise obsolete”, so it must have existed in Non-Austrian German as well, but I have never come across it before.
Both die Anerkenntnis und das Anerkenntnis are still with us. The neuter form and meaning is restricted to legal contexts, as with das Erkenntnis.
The shifting semantic sand dunes of “recognition” / “acknowledgement” / “admission” / “knowledge” / “guilt”. Eve’s lawyer was a snake.
From a DTA corpus search starting 1800:
Jean Paul: Titan. Bd. 1. Berlin, 1800.
— Unglücklich sind unsere jetzigen Jünglinge, die vom Baume des Erkenntnisses früher die Tropfen und die Käfer schütteln müssen als die Früchte.
four principal areas of pure mathematics
While I realize these are broad-brush divisions, they seem not to be exhaustive. There are analytic number theory and algebraic number theory, for example, but there is also a residue that seems to have no name (“elementary”?).
Those two are different: [algebraic number] theory and analytic [number theory]. The first is the study of algebraic numbers (there’s likewise also transcendental number theory), the second is the study of integers using the methods of analysis. Elementary [number theory] is indeed a thing, using, roughly, techniques older than analytical number theory.
I loved that stuff during my brief spell as a math major. It was applied math I hated.
Elementary number theory is very much a part of algebra (as mathematicians think of algebra). I actually found this rather disheartening, that it is nearly impossible to say anything interesting about, for instance, sums of powers of primes without introducing things like field extensions. (In a similar vein, proving the structure theorem for Abelian groups, “Every finitely generated Abelian group is a product of cyclic groups,” turns out to be an exercise in diagonalization of integer matrices.)
This study describes the development of an English and French multiple choice vocabulary test – the VocabLab tests – that measure learners’ knowledge at four frequency levels up to the most frequent 5,000 words: the 2,000-level, 3,000-level, 4,000-level, and the 5,000-level. The two tests aimed to address some of the limitations of tests currently in use. First, they are sampled from recent frequency lists. Second, they are geared towards Dutch-speaking learners of English and French in Flanders. Third, they attempt to minimize guessing by including an “I don’t know”-option. The findings showed that the tests are internally consistent. Mean scores decreased when the words were less frequent, lending evidence to the tests’ construct validity. Additionally, the tests seem to be able to discriminate between different proficiency levels. As both tests were developed according to the same principles, they can be used to compare learners’ English and French vocabulary knowledge.
The last statement is just a wishful thinking. The whole “number of words” concept is ill-defined and can be used only as a rough estimate.
I got an 89% for not recognizing 3 words (sniggle, apteral, fimbria), and for accepting two non-words. One of the nonwords was because of a simple mistake by stupid fingers, but the second looked like it “should” be a word: segulaation.
I recognized “segula” as a Hebrew word, although, truth to tell, I did not remember exactly what it meant.
So, segula. Huh. Live and learn.
While “segulaation” looks to me like it should mean “the process or act of making a segula“, it does not appear to be such a thing.
Second try, 92%. I missed the word “involucrate”, and wrongly accepted “pallman” (which looks like it should be a perfectly cromulent synonym of “pallbearer”), and “albedocentesis”. I recognized both parts of the term, and thought that “albedo” might be a rare term for the white of an egg, so that one could indeed do a “centesis” of it. Well, I was wrong. But “albedo” does also mean the white inner rind of a citrus fruit.
With aa, it would have to be Very Old Latin Indeed.
@Owlmirror: I have a somewhat funny story about the word albedo—although it’s not really my story, just something I saw on television.
I was watching the finals of the National Geography Bee on television once, about twenty years ago. That was the only time I watched, as I was a little curious what the competition was like. My elementary school started participating when I was in sixth grade, and I won the school contest and then qualified for the state competition. I did not do well at the state level, but I figured that I would have chances to improve, since the contest was four students in fourth through eighth grades. However, my middle school had never participated, and they weren’t about to start when I was a student there. (The middle school principal was evidently too involved in his Luftwaffe invisibility experiments to be interested.*)
So I had never seen what the Geography Bee competition was like at the national level. I had not been fond of the choices of questions at the state level, but the national rounds were, in many ways, worse. Almost all the questions involved identifying minor geographical features, ranging from little known to extremely obscure. The child participants (or the ones who did well, at least) must have prepared just by memorizing maps from all over the world. The final round came down to two tied contestants, and they kept going on and on, since they both kept getting every almost question right. In fact, they stayed tied through about ten questions, only one of which they both got wrong instead of right. However, that question was, of all the things they were asked, the one that was least a matter of pure trivia; it was something every actual geographer would know automatically—what physical geographers call the fraction of light reflected back by the Earth’s surface—the albedo. Yet neither one of the kids had the faintest clue; they had committed thousands of place names to memory, but they knew next to nothing about actual useful geographic concepts.
* Obviously, that joke requires some explanation. When the new principal took over halfway through my first year of middle school, one of my friends (who was very into Second World War stuff at the time) pointed out that the principal bore a strong resemblance to Hermann Göring. Moreover, sometimes during lunch, he would close the library to students, then stroll around the empty library apparently talking to himself. The joke became that Principal Göring was involved in a top-secret invisibility project for the war, and when he was seemingly alone in the library, he was actually briefing his cadre of invisible Luftwaffe pilots on their next mission.
if I can quickly come up with a plausible pronunciation then the word is probably real even if I don’t know it
I tried doing this test again, attempting to use this rule. I ended up with 100% real words and 30% (that is, nine) nonwords; two of which I really should have guessed because they looked like misspellings (advart, tolemization).
The rest were pleamishly; myllination; clomes, which Wiktionary says is actually a real word; pibby, smain, and oaning, which don’t happen to be English words (known to Wiktionary) but only by chance; and buggrist, which probably isn’t a real word but absolutely should be.
I’m happy to find out that I correctly guessed on gregarine, drover, and – this was a really out-there guess – embraceor. (“One who commits embracery.”)
embracery
I had to look that up. At first I thought it meant what this guy is famous for.
It turns out to be “the offence of influencing a jury illegally and corruptly.” (1808 J. Bentham Sc. Reform 72 To the same Jury not so well, on account of the danger or suspicion of embracery, and so forth.)
OED s.v. embracer:
Under “Spellings” they say “1500s–1800s embraceor.”
So entice meant ‘set on fire’, too. Second-oddest etymology I’ve learned recently, after atone < at one.
I heard an interesting word in this afternoon’s colloquium: bedo, meaning the fraction of incident solar radiation transmitted by the atmosphere—in other words, the quantity (1 − α), where α is the albedo. I’d never heard it before, but it clearly seemed to have been coined by interpreting al– as some kind of negating prefix and removing it. (Are there real examples of al– as a negator in English, or any other languages? I can’t think of any, but it’s entirely possible that I am missing some obvious examples that could have acted as the template for the backformation of bedo.)
For references, the speaker was Canadian, and she had some definite Canadian features to her speech. She used the commonly derided /ˈnu:kjələr/ pronunciation of nuclear, pronounced ruined as /ruːnd/, and referred to “a whole whack of things.” I didn’t quiz her about bedo, but I could later if I wanted to know more, since she is from another department here on campus.
There is a word bedo in the OED, but it’s just the transitive verb formed from the suffix be– and the verb do. It is listed as obsolete, but it has three relatively straightforward-looking senses, meaning: “put to,… shut,” “befoul, defile with ordure,” and “adorn, ornament, garnish.” A majority of the quotes are for the adjectival participle form bedone. I tried searching the Earth science literature for “bedo,” but I found nothing. It might be there, but the search results were overwhelmed by instances of the word “albedo,” either broken across a line break or scanned incorrectly. So if the term exists more broadly, it’s not very common.
I hope it doesn’t catch on. It seems dumb to me.
al- is a negator in Hebrew in some compounds, e.g. בֶּן אַלְמָוֶת ben almavet / ben ʔalmāweṯ ‘immortal’, or the modern אַלְחֶלֶד alxeled ‘non-rusting’, but that probably has nothing to do with it.
There should be a name for this phenomenon: folk etymologies/eggcorns based on imaginary morphemes.
Or it’s negative a-, leaving *lbedo which is reduced to bedo by phonotactics. But I’m just guessing, too.
If they had shown up in the Ghent test, I would have said that #s 2, 3 & 4 are words, and I would have nixed the rest.
That must have been the result of odylic forces operating on your cerebrum. (The etymology of odylic, which is a variant of odic is < Odin!)
Dimidiate is familiar to me from the asterisk-name of the binary numeric system, which should (by analogy with decimal) be called dimidial. I was also familiar with acatalectic. A catalectic line of verse is one that has a missing syllable in the last foot, like “Good King Wenceslas looked out / On the feast of Stephen”: an acatalectic line naturally is one with the normal number of syllables. It is obscure to me why we even need a word for a non-abnormal line, but there it is.
folk etymologies/eggcorns based on imaginary morphemes
I would say it’s just a variant of reanalysis: tamale is a morpheme in English with the regular plural tamales, but it is the latter that was borrowed from Spanish and reanalyzed; the regular Spanish singular tamal was not. So to those who knew Spanish, tamale was an imaginary morpheme.
Well, OK, I mean grammatical morphemes, which are a closed category.