RDIAENG.

Aoccdrnig to a rscheearch at an Elingsh uinervtisy, it deosn’t mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht frist and lsat ltteer is at the rghit pclae. The rset can be a toatl mses and you can sitll raed it wouthit porbelm. Tihs is bcuseae we do not raed ervey lteter by it slef but the wrod as a wlohe. ceehiro.


(Via Avva, who doesn’t provide a source, but the point is made without one.)
Addendum. Jim at UJG has a post on the subject in which he traces it back to what might be its source; check it out. (He also wrote about it back in May, using a different scrambled text; I don’t know how I missed it!)
Further addendum. A MetaFilter thread on the topic contains a remarkable example of the redundancy of language. The interior letters are not scrambled but completely changed… and yet the paragraph is still readable! Props to duckstab for thinking of it.
Yet another addendum. My record number of hits yesterday apparently came from a mention in a Slashdot post, where you can find further discussion of compression, entropy, and the like, as well as much silliness. (I was amused to see a comment titled “LOL languagehat misspelled iprmoetnt”; apparently Optic7 was under the impression I had written the garbled text myself.) I found out about the /. mention via Jim at UJG, who has a new post with further discussion of the matter.
But wait, there’s more… David Harris did an interesting experiment with this meme, which he describes in a recent post; the comments to the post include translations into French:

Sleon une édtue de l’Uvinertisé de Cmabrigde, l’odrre des ltteers dnas un mtos n’a pas d’ipmrotncae, la suele coshe ipmrotnate est que la pmeirère et la drenèire soit à la bnnoe pclae. Le rsete peut êrte dnas un dsérorde ttoal et vuos puoevz tujoruos lrie snas porlblème. C’est prace que le creaveu hmauin ne lit pas chuaqe ltetre elle-mmêe, mias le mot cmome un tuot.

And Portuguese (via the Merm):

De aorcdo com uma pqsieusa de uma uinrvesriddae ignlsea, não ipomtra em qaul odrem as lrteas de uma plravaa etãso, a úncia csioa iprotmatne é que a piremria e útmlia lrteas etejasm no lgaur crteo. O rseto pdoe ser uma ttaol bçguana que vcoê pdoe anida ler sem pobrlmea. Itso é poqrue nós não lmeos cdaa lrtea isladoa, mas a plravaa cmoo um tdoo.

Comments

  1. I posted this to my weblog yesterday with an attribution to David Harris’ Science and Literature weblog. David didn’t provide a source. I just did a Google search for Elingsh uinervtisy and found 224(!) hits, most dated today. It’s spreading like wildfire via email and blogs. It’s going to be tough finding out where it originated.

  2. amazing.

  3. fuckin rad!!

  4. This is really interesting, and something I’d like to know more about. It is close to a research interest of mine.
    Some questions come to mind:
    1) Does word length affect intelligibility? After all, the shorter a word, the fewer possible permutations for those few letters in the middle. In short words, maybe it’s only a choice out of two of three possibilities.
    2) It was easy to read your paragraph in part because the overall discourse context provided clues to the identity words. For example, syntactic clues tell you to expect a noun, a verb, etc. And you have the first and last letter. Suppose you are reading words in isolation? I bet it doesn’t work as well then.
    3) Very importantly, perhaps most importantly, it is crucial to consider how to measure the degree to which words are scrambled. A kind of metric could and should be devised. I guess you would calculate how far each letter had moved from its original position in the word. For example, if we took the word ‘steam’ and scrambled it to ‘seatm’:
    s is in situ = 0
    t has moved two places = 2
    e has moved one place = 1
    a has moved one place = 1
    m is in situ = 0
    Total movements from ‘steam’ to ‘seatm’ = 3
    As opposed to ‘saetm’:
    s is in situ = 0
    t has moved two places = 2
    e has moved one place = 1
    a has moved two places = 2
    m is in situ = 0
    Total moves from ‘steam’ to ‘saetm’ = 4
    Is ‘saetm’, with a larger index of displacement, harder to read than ‘seatm’? I’d really like to know more.
    –Tony

  5. Sdnous lkie smoe lgnisiuctis gard sdnetut has jsut been hdnaed a gaert tihses tpioc.

  6. I think a scrambled word with more vowels would be more difficult to read for instance eiere for eerie (eiree easier I think and has the same total index of displacement as eiere but one of the letters didn’t move at all which is something else to consider Tony. It’s also the mirror image(what is that called?) of eerie so maybe not the best example)

  7. An interesting point that turned up in the Avva thread is that it doesn’t work nearly as well in Russian (and presumably in other inflected languages), because the last letter is usually part of an inflectional ending and not of the base word, so that the only fixed meaning-bearing letter (so to speak) is the first, and the inflectional letters mixed in just add to the confusion.

  8. Yes, I think it works well in English because a lot of our words map to morphemes. So I wonder if words with affixes are tougher to suss out? I’ve got some more info on who may have down the original work over on my blog.

  9. It’s true, and an old finding (I remember learning that we recognize word shapes more than read letters back in cognitive psychology in 1996 or so). (Personally, I find it interesting that those words come out in my mental reading as a mouth full of dental implements, to my “ear”. Brain drawing an analogy?)
    I like using that fact, that readers tend to use shapes not letters, to smack about the head lo those idiots who say that the reason Europe ended up supreme over Asia is the versatility of the alphabet, or that character-based languages are impossible. Not that it addresses everything, but.

  10. In many languages the vowels are simply left out (Hebrew, Arabic, Persian). In classical Mongol (since 1206 A.D.!) the ideal writing system does not distinguish between o/u t/d and several other pairs, and cursive writing is still more vague, so that a/n can be confused at the beginning of words and u /umlaut-u /o / umlaut-o are not distinguished.
    Somehow classicism seems to thrive on inefficient writing systems, Chinese being the champ.
    “Thgruoh” seems to work for “through”, but I’m not sure that “tgrhuoh” would — “th” is one letter, sort of.

  11. A few days ago, this was posted to a mailing list I’m on which is populated by a lot of linguists, and they seemed to enjoy it. One, Jaques Guy, immediately posted with this counter-example:

    Hilpapy aoutrhs sitll issnit on fwinollog dullfready oslotebe snellpig.

    The longer, less frequently seen words add a lot of complexity to the information, and the scrambling of interior letters arranged so they form “fake words” makes it much more difficult to read quickly.
    Gabriel Landini also chimed in with the following thoughts

    Just note that in that text above there are many letters which seem to be in the same order [as] the original word. The other detail is that only words with more than 4 letters will be shuffled and in those 4 letter words, it is just a swap of 2nd and 3rd character only. By the length-rank Zipf’s law we know that common words tend to be short and therefore, they will be less shuffled that longer, less common words.

  12. It seems that the further along one gets in a sentence that’s been jumbled like this the easier it gets. Anticipating words to come later in the sentence being easier than at the beginning. Not sure. I also wonder what would happen comprehensionwise if you took a Chinese sentence and mixed up the tones, leaving the syllables intact? It would have to be spoken, not written, I suppose, or maybe written in Pinyin or Bopomofo.

  13. Let’s see if this pertains to painters of pantries done in loco parentis.

  14. Excuse me, the index of displacement for “seatm” and “saetm” are, of course, the same, namely 4 for both. My bad.
    –Tony

  15. My source was an unattributed post on the Lantra mailing list (it’s a huge mailing list where translators hang out). At the time I googled for the phrase and came up with nothing; two days letter it gathered a few dozen hits, none of them earlier than Sep 12, and now it’s in the hundreds. It _is_ spreading like wildfire, and probably it originated just a few days ago.
    See also this post at Jamie Jawinskie’s (of XEmacs and early Netscape fame), who wrote a small program to test the hypothesis and found out it works well (i.e. the counterexamples such as in Seth Morabito’s comment are the rare exception rather than the norm).

  16. This thing smells very urban legendy. It’s a joke, or at best a seriously watered down abstract of actual research done somewhere. Otherwise, why not name the researchers and the university? Also, if you fix the spelling, you’ll notice that the paragraph isn’t quite grammatical. The false attribution gives it the sort of weight (researchers, university) and timeliness (following the structure of typical news reports, the implication is that _new_ research results have been released) that writing “Did you ever notice that…” would not, so people bought it and blogged it.
    As for the claim: it shows some of the power of redundancy in language, but it’s not as simple as saying that letter order does not matter.
    What makes it work -– what makes it possible for you to easily scan the paragraph — is that you know the language on the sentence level and you have few words that fit into the blanks that make the sentence work. Structures such as “according to researchers at” are very common. You automatically fill in the word blanks — “research/researcher” (wwhich is it?) couples easily with “university” and each makes it easier for you to recognize the other and to recognize all that follows in the paragraph. The words you need for filling these blanks are not difficult to conjure.
    Also, it’s more than just the first and last letters that matter. Most of the letters have to be in approximately the right order and position to make it easy. It’s easy to make out “iprmoetnt” because “iprmoetnt” retains the right order, approximate location, and approximately correct sounds for p_r and t_nt.
    But as I said to begin with, this smells a lot like a “Didja ever notice…?” or “Isn’t it cool…?” remark disguised as a story of some supposed scientific novelty and import.

  17. Yes, it does have all the earmarks of folklore: the vagueness and the meme quotient (its rapid spread in three days across the Web). I’ve found two variants, one with “at an Elingsh uinervtisy” and the other with “at Cmabrigde Uinervtisy”. There was a similar jumbled text that made the rounds back in April/May which at least mentions a publication and a researcher: New Scientist and Professor Kourosh Saberi. It reminds me a little of the Mark Twain English spelling reform story that has resurfaced recently as an EU language policy joke.

  18. By the way, try this this if you’re interested in making similarly scrambled texts.

  19. I suppose this is proof positive of the validity of “whole language” over phonics. :^)

  20. I suspect that the distribution of ascenders and descenders in the words plays a major role in readability, as does the density of the letters.
    Compare:
    (1) sbalermcd
    (2) searclbmd
    (3) saremlbcd
    In (1), the two ascenders are way out of place. In (2) the ascenders are in the right place (though switched) but the dense “m” and lightweight “r” are still wrong, producing a different pattern of light and dark as your eye scans the page. (I’m typing this in Courier– your mileage may vary.) In (3) this problem has been fixed. I’d predict that as you go down this list, the words become easier to read.
    Now where’s that psych/linguistics grad student?
    Someone should test this out!
    Cassidy

  21. I don’t agree with your theory, but I love your site!

  22. Thanks! I like your blog too.
    Just curious: why don’t you agree with that theory?
    –Cassidy

  23. It would be interesting to see how a prefix affects the readibility of an otherwise obvious word. such as “dsiturtibe” versus “ruisdribttee”
    while there may be an urban legend element to this whole thing, it is also hard to deny that it is surprizingly easy to read text scrabmled in such a way. On the other hand, many words can be dropped from a sentence without affecting it’s intelligibility. perhaps you somehow concentrate on creating permutations of the words that are most likely to be important, and once you have one third of the words you could simply fill in the rest of the text with ONLY the first and last letters.
    To conclude, I really wish someone would run a good experiment on this.

  24. “By the way, try this this if you’re interested in making similarly scrambled texts.”
    So cool and yet so useless ;)
    Just kidding, its actually quite fascinating.

  25. John Carter says:

    Just tried scrambling the scrambled text five times in a row…
    Adcnocirg to a resarchceh at an Ensilgh uvteirnisy, it den’sot mtater in waht oedrr the ltetres in a word are the olny iepnrmtot tinhg is taht fsrit and last lteetr is at the rghit pclae. The rest can be a tatol mess and you can stlil read it wothiut peorblm. Tihs is bceause we do not raed ervey leettr by it self but the word as a wlohe. chreieo

  26. Doesn’t seem to affect it much, does it?

  27. How about if you do all caps:
    ADCNOCIRG TO A RESARCHCEH AT AN ENSILGH UVTEIRNISY, IT DEN’SOT MTATER IN WAHT OEDRR THE LTETRES IN A WORD ARE THE OLNY IEPRNMTOT TINHG IS TAHT FSRIT AND LAST LTEETR IS AT THE RGHIT PCLAE. THE REST CAN BE A TATOL MESS AND YOU CAN STLIL READ IT WOTHIUT PEORBLM. TIHS IS BCEASUE WE DO NOT RAED ERVEY LEETTR BY IT SELF BUT THE WORD AS A WLOHE.
    CHREIEO

  28. I think much of it has to do with context. I can make out simple substitution codes based on context fairly easily, whereas random separate words would require a statistical approach. If you expect a word to come next and something similar to it does, you’ll be able to interpret it.

  29. Robert Schwartz says:

    Would it work in Hebrew?
    http://volokh.com/2003_08_24_volokh_archive.html#106205693334854073
    Do different languages use different amounts of redundancy in their orthography?

  30. Matthew Kramer says:

    This seems to have some applications for privacy. If the government is trying to filter e-mail or files or newsgroups for certain “bad” words, if you make your posting after putting it through a scrambler that mixes things up like this (except for the first and last letters) it would make it a lot harder to find the “bad” words, but your message could still get through.
    Also, can a scrambled message be “obscene?”

  31. hi, i’m an english language n literature student at a university in singapore. i’m rather interested in doing a study on this scrambled letters theory. but first of all, i think this blog’s great. everyone has had some useful thoughts and ideas. brilliant! btw i jus happened to surf in whn i googled ‘Elingsh uinervtisy’. as for now, i’ve got a language class to go to, so ciao! wil b back soon…! in the meantime, food for thought: wil this make it more difficult to read?
    AdCnOcIrG tO a ReSaRcHcEh At An EnSiLgH uVtEiRnIsY, iT dOn’SoT mTaTeR iN wAhT oEdRr tHe LtEtReS iN a WrOd ArE tHe oLnY iEpRnMtOt tInHg iS tAhT fSrIt aNd lSaT lTeEtR iS aT tHe rGhIt pClAe. ThE rSet cAn bE a TaToL mEsS aNd YoU cAn StLiL rAeD iT WoThIuT pEoRbLm. TiHs Is BcEaSuE wE dO nOt rAeD eRvEy LeEtTr By It SlEf BuT tHe wRoD aS a wLoHe.
    ChEeRs

  32. The text with middle letters totally replaced is decidedly hard to read, and I fail at some places (but then, English is not my native tongue). The capitalizations slows down reading, but does not make it unintelligible.
    If we rseicrtt oruelesvs to smbrainlg jsut the lterets in ehac slylalbe, does it beomce sginifinactly eaeisr?
    Interesting…

  33. Mathowie said at MetaFilter:
    It’s a well-documented phenomena in cognitive psychology. People recognize words by their shape. Ever notice how much easier it is to read (recognize) lower case text? SEE HOW MUCH HARDER IT IS TO READ TEXT IN ALL CAPS? IT’S BECAUSE WORDS LOOK LIKE FEATURELESS BLOCKS instead of sporting nice ascenders and descenders to give you a clue to what letterforms words contain.
    This is also why the really important bits of all legal documents are in small, capitalized text: it’s purposely done to reduce reading comprehension so you won’t read the contract.
    But is it not also true that we sometimes capitalize words to make them stand out, so that people WILL read them, or to emphasize on these words?

  34. As someone else has said, this is fairly close to research interest of mine as well.
    Most of this seems fairly close to the concept of non-words and orthographic non-word neighborhood effects (i.e., RT’s in a yes/no word/non-word lexical task being longer when non-words have a higher neighborhood, or lexical inhibition). Since most of these are strictly *non-words*, I’m betting that the non words with higher orthographic neighborhoods are harder to read.
    Also, since we’re talking about representing these non-words as actual words I’d be willing to bet there are two very obvious factors involved. I’m betting the base-words (non scrambled) with a higher orthographic neighborhood are even harder to recognize, e.g. high non-word orthographic neighborhood, high base-word orthographic neighborhood = much more difficult to recognize.
    And, as someone pointed out, we’re not even considering ascenders, descenders, syntax, schemas for written text, the degree to which each word is scrambled….
    But that’s why I love this stuff.
    S.W.

  35. I’ll just chip in some thoughts regarding Tony’s comment:
    1) Does word length affect intelligibility?
    Long scrambled words compared to long, but correctly spelled words? Probably not.
    You should look up some textbooks on cognitive psychology. It’s well documented that the spaces between words are very important for reading. (Remove the spaces or replace he spaces with any letter, for example “x” and try reading.)
    “AoccdrnigtoarscheearchatanElingshuinervtisy,itdeosn’tmttaerinwahtoredrtheltteersinawrodare”
    The saccadic eye movement vary when reading and depends on the length of the words. Long words will cause longer movements. Probably explains why the letters at the start and end of a word are very important for perceiving the word.

  36. @ S.W. The effect you mention is probably one of decision making (=a strategy effect) in the Lexical Decision Task (LDT), not relevant to visual word recognition per se. The MROM-p (Jacobs et al 98) for instance uses a threshold mechanism for deciding word/nonword responses.
    Apart from that, neighbourhood effects depend on frequency. Higher frequency neighbours hinder, lower freq neighbours help. That said, nonword (scrambled word) neighbourhood effects certainly influence difficulty, but one would expect that the effect depends on the frequency of the respective neighbours versus the frequency of the base word. Lower frequency nonword neighbours might actually help in that the push the same letters but loose against the higher frequent base word when it comes to word activation (at least it’s that way in the MROM-p, Coltheart’s DualRouteCascaded Model, 2001 should have some problems accounting for this, since neither direct lexical access nor grapheme-phoneme conversion will help in this case IMO). Higher frequency nonword neighbours almost certainly hurt recognition.
    That said, I think base word frequency is more important than neighbourhood.
    more in comments here: http://www.crookedtimber.org/archives/000525.html

  37. As stated before, it is a well known phenomena. I don’t even study something in that vein, but I know, that people read words not letter by letter. That’s why you can read a book very fast.
    But what I think is really interesting, is the way, you could teach children with that sort of method. Or if it is better to improve a forgein language with that mehtod. It is still easy to read an english text, but you put a little more effort in getting it right, when reading the “deciphered” text. (I’m talking about me, as you already noticed, my English isn’t very well either!) It took me twice for some words (also they were easy) to see if they were right.
    BTW: Whar abour all those script kiddies on IRC or on thousands of Online Games. I tend to misspell words very often, because the layout of my keyboard is pretty stupid and sometimes several fingers are too fast! ;)

  38. @Daphne
    what mathowie said is likely wrong. You’re right about caps, though one might argue that increasing reading difficulty increases time spent on focussing the word, and if the eye=mind assumption holds that would mean more thought spent on the word, which in turn basically is equal to emphasis.
    Concerning mathowie, I made the case against his view at the crooked timber post from my previous entry. Short version: there’s an advantage to “normal” word form (as opposed to all-caps) due to perceptual frequency, but there is no shape information stored in the brain as such.

  39. @thomas: the “Direct lexical access” hypothesis, otherwise known as “whole-word reading” is not exactly universally accepted with in the field. Max Coltheart (DRC, see above) pushed it, but most of the evidence in favour comes from the study of aphasic and alectic patients and is not really convincing IMO. Apart from that, successful models have been built without Coltheart’s two routes (lexical access & grapheme phoneme conversion) which can account for an equal number of findings. The most recent Coltheart model incorporates a parallel-distributed processing model (some of this type (e.g. Plaut, McClelland, Seidenberg & Patterson 1998) have been very successful on their own) and as such does not really do whole word lexical access anymore.
    The popularity of the Coltheart account stems from a)agressive marketing and b)simplicity. In the beginning it was a model of either look it up in a dictonary (lexical route) or piece it together (grapheme-phoneme conversion). Clean, simple, with intuitive appeal. Which is why it is/was spoon-fed to teachers and others outside the field. Much easier than explaining a neural network to them.
    These days the model lost it’s simplicity, effectively admitting that there are no “rules” in the brain and that you don’t need a “mental lexicon” because distributed representation is sufficient.
    Concerning teaching, yes it makes sense to teach whole words (for various reasons), but in other languages (e.g. Spanish) you’re better off teaching syllables. The determining factor is something you might call the grain size of lexical processing. For highly regular languages (e.g. German), that have a more or less one-way mapping of graphemes (letters) to phonemes (sounds) your better off learning to speak by letter, for intermediate languages like Spanish you stick to the syllable and for English (which is a mess) it makes sense to stick to words because it is so irregular. As to French, which is even worse, it’s inconceivable to me how anyone can learn the language at all ;).

  40. Dyslexia?

  41. Cassidy: Because your examples didn’t work for me (I found them all about equally easy to read). Not saying you’re wrong; as you say, it should be tested.
    Daphne: The two things aren’t incompatible; we can put things in caps to make people read them, and they can take longer to read (which may actually be a good thing if you want people to pay attention to them).
    Everyone: Thanks for the great comments!

  42. @thomas: yep, deep versus shallow dyslexia, that’s Coltheart’s main line. I did have to learn most of that stuff for graduation exams, but managed to forget most of it right afterwards. Anyway, it turns out that most “pure” cases are not so pure after all, more subtle tests show there is no clear cut dissociation, and so the basis for the distinction falls apart.

  43. @Markus: Thanks! :]

  44. LH: I also saw your name spelled Language Chat on one tracked back site. Like drinking tea out of the tip of a czapka.

  45. Above someone posted a scrambled message in which they used the word: “smbrainlg”
    I assume this was supposed to mean “scrambling”, but because it was missing a ‘c’ I had to stop and read it over again several times to realize it was just misspelled. So while the order of the letters might not be important, the existance of all of them and in the right quantity seems to be!

  46. holy crap. this is the best shit since sliced meat.
    \m/ -_-

  47. Hi Language Hat! Did you know you’ve made it to Snoeps?
    http://www.snopes.com/language/apocryph/cambridge.asp
    See you in Pepys’s dairy!

  48. However in the case of “Stana” it could work out as Santa or Satan, and could become a case for words being also contextually understood.
    It’d be pretty easy to discern within
    “Heer cmeos Stana Calus, hree cemos Staan Clsau, rgtih dwno Stnaa Cusua lnae…”
    but perhaps things would would be more often misunderstood when written as, “Mnay chrliend wpshpori Saatn on hghi hdlioays.”
    I’ll notify the PTA.

  49. These are typewritten scrambled words which may prove easier to recognize and read at our normal speed. I doubt most people will read with the same ease if the words were handwritten.

  50. Thanks for all that, everybody. I wrote a manuscript about Spelling and Society ten years ago but no publisher would look at it because they all said nobody except specialists would buy a book about spelling that was not about how to learn to spell correctly.
    But spelling is actually fascinating, as you know.
    I would like to ask you to take your interest further – what sort improvements in spelling for English would make it more user frendly for lerners, without upsetting people who can read alredy, and still leaving everything in print now easy to read?
    You can read scrambled words easily because:
    • you know enough of the English language to be able to work out words from their context. There was/is a fad in schools for making kids do ‘cloze’ exercises, that is, one word in ten or so would be cut out, and if the children could put in the word, that meant they were reading with comprehension. eg
    How my SPELLCHECKER can read that scrambled word paragraf. The only words it could not read are in capitals -
    : According to a RSCHEEARCH at an ELINGSH UINERVTISY, it doesn’t matter in what order the letters in a word are, the only IPRMOETNT thing is that first and last letter is at the right PCLAE. The rest can be a total mess and you can still read it WOUTHIT problem. This is BCUSEAE we do not read ERVEY letter by it self but the word as a whole. cheerio.
    My spellcheck here has not been looking at word shape – it has been registering all the letters and doing a sort of little shake in its mind to come up with real words that are in its dictionary.
    And the same with us in reading words – or writing them with typos, often. We’ve got the letters for the words in our minds, but it may be an extra step to get them in the right order for reading or writing.
    So we still have the alfabetic principle there, mixed up with other principles.
    Word shape is only a very small part of identification – eg with ascenders and descenders and the way that the bottom half of words gives more information than the top half – but even relative learners dont usually get fazed with different shapes and sizes of fonts. As long as there is not too much variety within words.
    A proof that it is not entirely word shape that enables you to read jumbld text is that the jumble alters the word shape, does it not. All that’s left in place are the first and last letters.
    There’s a lot of reserch about the fact that readers rely on these more than they rely on the middle letters when reading – but the better the reader, the more they are able to use multiple information in reading with speed and accuracy. While auditory perception can be the greater back-up in decoding and learning for example, visual recognition weighs in with practice and words become familiar. And it is not auditory perception in ‘real’ sense time for fast readers either. They also use auditory perception to hold sentences in short-term memory, so they tend to be more sensitive to ‘rhythm’ as style in text than poorer readers, because the ‘metre’ of sentences helps this ‘auditory’ memory. (Some people usue auditory memory for telefone numbers, some use visual, and some, like me, use both, plus a mnemonic such as the date of the Battle of Trafalgar. People usually have to use visual memory to get English spelling right – ‘does it LOOK right?’ because spellilng doesnt always follow any fonemic sense. I use auditory – sounding out the letters if the spelling is silly – and just reason if it is not silly – and these strategies made me a perfect speller.)
    What happens is that readers quickly get a ‘set’ for the type of text they are reading – the language it’s in (if they are multilingual), the font it is in, and – the sort of spelling it is in.
    If you are reading ordinary text and you suddenly have scrambled words – or even spelling mistakes – it disrupts you far more than if the whole sentence is scrambled words. If upper and lower case are mixed in words, the words take longer to read.
    A similar example is in speaking or writing sentences – if I (I dont know about you) write (curious about auditory influences, I wrote ‘right’ first) out ideas as I think of them, they do not come in correct English syntax. One reason why we do not write as quickly as we can think is that our minds have to sort out the ‘correct’ grammar for communication. James Joyce realised this in Ulysses – we often even have to get the thoughts themselves in some sort of order because the stream of consciousness is not that good at thinking straight either
    Look also at what mobile fones are getting up to with text mesajes – the minicomputer uses probabilities to work out what words are when you only enter key letters which are on your keyboard grouped with other letters for economy.
    Now I am going to send you an email about the problems that most people have lerning to read. You lot are lucky ones who did not have that much dificulty – but most peple do not even read well, and would have problems with ordinary text, let alone scrambled. I have a simple experiment that can prove that most people cant read well.
    I have another simple experiment of sixteen words that proves that very few people can spell very well either – even if they are teachers of reading or psychologists specialising in dyslexia or intelligence. The reason is not their fault – it is just that the English spelling task is too silly.

  51. This 1999 comment in the alt.2600 newsgroup indicates that the original text appeared as a reader’s letter in New Scientist…
    http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&oe=UTF-8&selm=375AF032.438C%40dante.control.systems

  52. Roger: Thanks! Here‘s the direct link.

  53. Hmmm. That may have a tie-in with another NS article concerning structure and content of communications. If it’s true that written English has a Shannon entropy of 8 or 9, coupled with the possibility that an initial recognition of a word *can be* (important distinction) dependent on the first and last letters along with the context in which the word is being read, then that might explain how a speed reader would only notice some 3-4 errors in an A4 page of jumbled text (as quoted by the original letter writer in 1999, referring to his PhD thesis of 1976).

  54. As a followup, the NS article is “Look Who’s Talking”, New Scientist vol 179 issue 2403 – 12 July 2003, page 36, if anyone’s interested…:)

  55. And, for what it’s worth, my exposure to the meme came as a result of a colleague sending me a copy which referred to “Oxofrd” as the university, which may make it yet another variant:
    A Birllaint Obsrevatoin…
    Aoccdrnig to extnesvie rseeacrh conudcetd at Oxofrd Uinervtisy in Enlgnad, it deosn’t raelly mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht the frist and lsat ltteer is at the rghit pclae.
    The rset can be in a toatl mses and you usulaly can sitll raed it wouthit much porbelm. Tihs is bcuseae we do not raed ervey lteter by it slef but
    the wrod as a wlohe.
    Jsut thnik a momnet abuot all the tmie you and I watesed laernnig how to splel wrods croreclty!

  56. The meme mutates!
    Received via my friend Lawrence Kesteloot from his aunt in Belgium:
    “Sleon une édtue de l’Uvinertisé de Cmabrigde, l’odrre des ltteers dnas un mot n’a pas d’ipmrotncae, la suele coshe ipmrotnate est que la pmeirère et la drenèire soit à la bnnoe pclae. Le rsete peut êrte dnas un dsérorde ttoal et vuos puoevz tujoruos lrie snas porlblème. C’est prace que le creaveu hmauin ne lit pas chuaqe ltetre elle-mmêe, mias le mot cmome un tuot.
    “Mdere aorls !!”
    Any native French speakers care to comment?
    –Cassidy

  57. Harold Angel says:

    I found the University research fascinating.
    However, although my wife and I had no difficulty reading the text, both my children who are dyslexic found it very difficult to read. Has it been tested on various structures of society or only on University students.
    Harold

  58. It doesn’t work in Hebrew.
    Since Hebrew is mostly written without almost any vowels, the meaning of the words are “collected” from the words surrounding them (even words written later in the sentence).
    Mixing the letters in Hebrew words makes reading much harder because several words have to be de-mixed at once.

  59. As far as the French version goes, I’m no native speaker but I managed (mostly) OK, and when you consider that I haven’t even read French prose in probably decades that’s quite an achievement. In fact, “Mdere aorls!!” leaped out at me, so Mdere aorls indeed :)

  60. It works in Bahasa Melayu ( Malay Language ) :
    meunrut kpedaa stau kjaian sbeauh unvivrestii igngiers aurtan hruuf di dlaam stau pretkaan adalah tadik pnetnig, prkeraa umtaa ilaah hruuf premtaa dan trkaeihr stieap prkeatan msetliah di psoisi ynag butel. Ynag lian-lian bleoh dicpmuaraudkkkan dan sesoernag misah bloeh mmebcaa pretkaan tresubet tnpaa kermuitan.

  61. Though I’m a native English speaker, I first heard about this on a French website’s games forum where I saw the translation in French and was surprised to find that I could easily read what it said.
    I thought it was a joke at first but then I discovered your site via a link from sci.lang.translation and everything became clear.

  62. Mes cinoseus animet ragreedr la tioslieven
    This is a better test of applicability in a second language (French) for the native English speakers who have been reading the material on this website. My guess is that anyone with even a passing knowledge of French could read the original scrambled French passage (posted earlier) quite well because it is a direct translation of the same thing that you had already seen in scrambled English. If your French is less than perfect, then recognizing familiar words is not significantly more difficult when the words are spelled incorrectly. Even when looking at the letters in scrambled passage, you might not know what the proper spelling is. Instead, you know that there is a word that looks like the scrambled word, and that is sufficient for comprehension. The key factor, then, is context.
    For the sentence “Mes cousines aiment regarder la television”, there is no context that suggests what the passage is about. My hypothesis is that native and fluent French speakers will read it fairly easily, but anglophones who know some French will have trouble with it.
    What does this mean in general? Being literate, which includes knowing how to spell, helps you to recognize letters as words — even misspelled words, as demonstrated by the research. Those who are not fully literate can “read” words when the context is clear (e.g., road signs, choices on a menu) but otherwise struggle to understand written language because letters are seen to represent ideas rather than words.
    If that is obvious, forgive me. I am a chemist with no knowledge of linguistics or cognitive psychology. I am writing, though, to make a case against the notion of whole-language reading, as some might interpret the research results to conclude that “spelling doesn’t matter”. My mother was a special-education teacher at the grade school level, and she saw that the shift away from teaching phonics was a great disservice to schoolchildren. She knew that her job was secure because the use of whole-language reading instruction would always result in students who needed to see her in order to learn how to read.

  63. I think that’s a good point, and it did take me significantly longer to figure out your sentence than the earlier passage, though this probably has as much to do with the longer words as with the lack of context; the first word I got was “animet,” the shortest of the scrambled words.

  64. You might find the reaction from the artificial intelligence research people interesting. The implications to natural language recognition that can handle all of the typos and misspellings humans are prone to is significant.
    Here is a page about a computer program that uses a similar approach to the one described in the original scrambled text. There is even a link to a free demo program does this.
    http://www.ai-forum.org/topic.asp?forum_id=1&topic_id=9528

  65. Eve Garrick says:

    I received this over the instant messanger from a friend, but I’m glad I now have a few sources to back it up in my “neat things” file.
    I noticed someone mentioned the ALLCAPS problem and the VaRiAbLe caps and lowercase. I didn’t have any problem at all reading this, though it might be from the paragraph having sunk into my brain.
    But then again, I did some research on hyperlexics. Since someone mentioned dyslexia, how do you think this paragraph would do with someone with the almost opposite syndrome of hyperlexia?
    (mostly my interest is because, while I was never formally diagnosed and I’m much too old to have it done now, I was an early reader. A good example of hyperlexics in action is seeing how well you can read upside down. I can breeze right through an inverted or backwards text. My roommate can’t but my techie computer friends (who also said they read early) can. Neat.

  66. This is the first I’ve heard of hyperlexia, but it sounds like a pretty useful syndrome to have!

  67. So we have (what we could henceforth call) the “msesgae”:
    “Aoccdrnig to a rscheearch at an Elingsh uinervtisy, it deosn’t mttaer in waht oredr the ltteers in a wrod are, the olny iprmoetnt tihng is taht frist and lsat ltteer is at the rghit pclae. The rset can be a toatl mses and you can sitll raed it wouthit porbelm. Tihs is bcuseae we do not raed ervey lteter by it slef but the wrod as a wlohe”.
    Even if it’s funny, this “msesgae” is an improper and excessive generalization, which conveys an extremely reductive vision. Moreover, whereas it should only remain what it is, i.e. a simple fantasist and entertaining text, it is taking worrying forms (we see it in mails, weblogs, chat-rooms where participants, absolutely amazed and amused, are venerating this “sensational discovery” and friends from everywhere (also excited) are forwarding it in different languages (apparently, this “hoaxmeme” (hoax + meme) is floating all over the web).
    Let’s try to encircle the topic (not by haughty pedantry but just by anticonformism and anti-“simplistism”). If you were looking for a serious explanation of it, here is an “anti-hoaxmeme”:
    Introduction
    Reading is a complex activity that involves many aspects of knowledge, which are of various natures and various complexities (this is due besides to the fact that “writing” is complex). It’s an activity, which implies cognitive processes but also, simultaneously, perceptive processes: reading, it’s to perceive and to identify words.
    Development
    Many linguists worked on the description of the mechanisms’ evolution of the words’ identification and there are now many developmental models of reading. The principal models comprise three way of reading, which correspond actually to three chronological stages of acquisition (for this presentation, let’s start with the second one):
    - the alphabetical reading (second stage): the reader connects the oral examination with the writing (in other words, he learns how to make correspondence between letters and sounds (ex: the sound [k]can be written with ‘c’ (cot), ‘k’ (kiss) or ‘ch’ (chord)). At this stage of phonological mediation, there is a code training; the learner enriches its phonological knowledge and transfers it to new words (it’s a form of self-training). This stage is called an “indirect way” because the reader reads the words through a decoding process.
    - the orthographical reading (third stage): the words are analyzed in orthographical units (orthography indicates here the sequence of letters forming the word). There is no phonological conversion; the words are read and recognized directly in reference to a memorized orthographical lexicon. This stage replaces gradually (but not entirely) the alphabetical one. The reader does not need to decipher anymore: he recognizes the words through a “direct way”.
    - the logographic reading (which is actually the FIRST stage in the reading training): at this stage, the reader uses various kinds of clues to ‘read’ the words, inter alia, those provided by the extralinguistic environment. The letters’ order and the phonological factors are not taken in account, but the visual clues are. There can be at this stage an instantaneous recognition of familiar words (or somehow ‘learned by heart’), and the riddles made on the basis of projecting visual clues allow the constitution of a first total vocabulary. The visual clues can simply be the length of the word or its “silhouette” (outline) or even just one letter. The classic example to illustrate this stage is the word: “Coca-Cola”, of which logo is easily identified by almost all children of 5-6 years old. If we change only one letter of the word: “Coca-Coca”, children will not notice the difference from the original word (adults neither sometimes, as some experiments proved it).
    The most perspicacious of you may have already understood: what occurs actually when we read the “msesgae”, it is that we, literate readers to whom reading and writing have been taught, use our competences, acquired and automated thanks to years of reading experience. In other words, we have developed “HABITS” of reading.
    The “msesgae” experiment could let us think that we get back to a logographic reading, in which access to significance is carried out directly via the pictorial semantic system (with words treated like images-logos), but this is not completely true.
    Actually, we continue to use the orthographical reading system (in which access to significance is carried out via the verbal semantic system). If we look at the “msesgae » more closely, we can notice that 34 of its 68 words (short and common by the way), are correctly spelled (50%, half of the text, and most of them are “grammatical words”). Added to a simple and common syntax (journalistic style of the “forma brevis”) and our capacity of anticipation and auto-reflex correction of more or less experienced reader (the system used is close to the “typing error” one, and anyway, teachers manage quite well to read our essays stuffed with spelling mistakes. In other words, you don’t have to be a Professor of literature to spot “what” in ” waht “!!!), it gives many visual clues!!! (Moreover, there is a syllabic facilitation phenomenon, but I skip the details).
    Conclusion
    The proposition, which is conveyed through the «msesgae», is not completely false but it is very reductive, and completely incorrect when it affirms that only the place of the first and the last letter of the words do matter. Actually, it deals more with their “silhouette” (from which our (almost standard) system of abbreviations rises (another facilitating clue)). If we can read the “msesgae” without any problem, it is because we are good readers reading a text easily accessible in spite of its orthographic and spelling mistakes.
    To prove it, if I give you the correctly spelled words “acetoxybutynylbithiophene deacetylase” or “carboxymethylenebutenolidase”, dear expert readers, you will resort to an alphabetical analysis (second stage) and will use a grapho-phonological decoding for these unknown words (I suppose, this experiment may not always work if you are chemist, druggist or doctor… if it’s the case, sorry for this affront :-).
    Another counterexample: if you read AT THE FIRST GO the following sentence as quickly and fluently as you did with the “msesgae”, all my theoric explanation goes down the drain (or you are an innate champion of anagrams!):
    “Nreuuoms pmeeononnhs peossss uiapocmltecnd etaaoilxnpn; nwttdtsniinoahg, the pdseuo-snfiiiectc spssliiimm is not snfiiiectc and eieecndvs are oetfn mdanleiisg”*.
    Guillaume Fon Sing,
    (alias GUITCHUS)
    guitchus@hotmail.com
    Linguist
    * “Numerous phenomenons possess uncomplicated explanation; notwithstanding, the pseudo-scientific simplisism is not scientific and evidences are often misleading”.
    Please forward it, …it can teach sb a thing or two.

  68. Tihs eltuilischansaty ecbermad trehoy taht all you need is the fsrit and lsat lteetrs of a wrod to be albe to raed it wkros mcuh of the tmie, but has its laoittimins, eacellipsy wtih lnog wrdos. Can you tlel waht aoboirghapa manes, for eplamxe?

  69. No, but that’s because of lack of context. I could make out the rest, even the long ones, because it fit into the sentence pattern. The point isn’t that it makes no difference whether letters are scrambled, but that there’s enough information left—in conjunction with contextual clues—that you can make out the words anyway, with greater or lesser difficulty depending on how long and obscure the words are. I don’t know why so many people have such a problem with this; it seems pretty straightforward.

  70. Tom Whitmore says:

    Partially crossposting here and on MAKING LIGHT — “agoraphobia” leapt out at me instantaneously, but I’ve been playing with anagrams for a while now. I’m wondering whether the ability to “see” instantly typographical errors on a page is related to this concept. Is this a form of hyperlexia? Does it correlate with the ability to read upside-down/inverted text? Does anyone else do a double-reflection on the inverted “ambulance” on the front of an emergency vehicle seen in one’s rear-view mirror (I know it’s inverted, so I have to invert it and then I invert it again because it looks wrong — there’s a very slight mental hesitation with each inversion, noticeable when they’re chained).
    Cheers,
    Tom W.

  71. Chuck Nolan says:

    C E I I I N O S S S T T U V

  72. There are some amusing Spanish variants at juegos de ingenio (under “22 de septiembre”; no permalinks).

  73. My husband is dislexic and has a hard time reading and his spelling is terrible. I on the other hand notice misspelled words right away. I was able to read the paragraph but I stumbled over some of the words. So I thought let my husband try and read it see what happens. He read it faster than I have ever heard him read. This was very interesting to me.

  74. Stan Telega says:

    Think of when you are reading a book and you finish the sentence and say to yourself “This doesn’t quite make sense” so you go and reread and sure enough not only did you misread a word but as fast as you read you misread it with another word that had meaning, had the exact number of letters (or within one letter) and you did that immediately. Is that because our brains find meanings everywhere for everything ?
    I recommend THE LANGUAGE INSTINCT by Stephan Pinker subtitled How the Mind Creates Language.

  75. As a large part of my job includes copyediting, this conjures numerous nightmares. One technique I use for small documents is to read them backwards. That way, word order doesn’t kick in and cause my brain to “correct” what I am reading. A mistake as sipmle as this can cost thousands of dollars when something has to be reprinted. Reading backwards forces you to look at words individually and out of context.

  76. Interesting. I constructed each word over 4 characters in that paragraph quite carefully – for the longer words, sorting the letters into alphabetical order before re-sorting them so that no pair of letters was the same as in the original word. When a word appeared twice, I mixed the letters differently so that the same pattern didn’t appear. I know I could have worked out “agoraphobia” – but not immediately. (I’m not sure I would have known “enthusiastically” very fast, though context would have made that easier.) I would have had to sort out the letters mentally like playing Scrabble, moving them around in my head to figure out what word they could make.
    When I play Scrabble, I get two and three-way words with reasonable ease: I was amused to discover that I could do it playing numeric Scrabble, too, suggesting that it wasn’t directly to do with my language skills.

  77. Third year grad student in painting and drawing considering this topic for thesis: Can portraiture be free of all forms of representation of the physical appearance (Cmabrigde Uinervtisy) without loosing the intentional sign of the individual (word)? I’m looking for documentation. Can anybody help?

  78. If anybody’s still reading this thread, go check out the comments at Making Light; there’s some very interesting stuff (as well as a great anecdote about a Chinese restaurant).

  79. Here’s a new twist to the “research” that’s doing the rounds…!
    Iltnsegnetiry I’m sdutynig tihs crsrootaivnel pnoheenmon at the Dptmnearet of Liuniigctss at Absytrytewh Uivsreitny and my exartrnairdoy doisiervecs waleoetderhlhy cndairotct the picsbeliud fdnngiis rrgdinaeg the rtlvaeie dfuictlify of ialtnstny ttalrisanng sentences. My rsceeerhars deplveeod a cnionevent ctnoiaptorn at hnasoa/tw.nartswdbvweos/utrtek:p./il taht dosnatterems that the hhpsteyios uuiqelny wrtaarns criieltidby if the aoussmpitn that the prreoecandpne of your wrods is not eendetxd is uueniqtolnabse. Aoilegpos for aidnoptg a cdocianorttry vwpiienot but, ttoheliacrley spkeaing, lgitehnneng the words can mnartafucue an iocnuurgons samenttet that is vlrtiauly isbpilechmoenrne.
    Or, if you prefer:
    Interestingly I’m studying this controversial phenomenon at the Department of Linguistics at Aberystwyth University and my extraordinary discoveries wholeheartedly contradict the publicised findings regarding the relative difficulty of instantly translating sentences. My researchers developed a convenient contraption at http://www.aardvarkbusiness.net/tool that demonstrates that the hypothesis uniquely warrants credibility if the assumption that the preponderance of your words is not extended is unquestionable. Apologies for adopting a contradictory viewpoint but, theoretically speaking, lengthening the words can manufacture an incongruous statement that is virtually incomprehensible.
    LOL… it gets worse!

  80. I’ve been thinking about what makes super-fast reading possible (incidentally, this phenomenon is NOT hyperlexia, which involves a deep concentration on certain sorts of words that sounds vaguely similar to Asperger’s, and an inability to communicate well verbally).
    My new office-mate is dyslexic. Without outside help (she has a reading program on her computer) she reads about 5 pages an hour. I read around 60 pages an hour; if intentionally speed-reading, I can read at least twice that. I know that this is somewhat unusual – when I took the speed-reading course in college, the instructor was pretty much blown away that I normally read faster than he had predicted we’d be able to read AFTER taking the class.
    My question is this – I got about half-way through the garbled message before realizing that it was, in fact, garbled. Has this happened to others? What might it mean?
    I’m awful at spelling, too, I think for the same reason (i.e. I’m simply “recognizing” words as I go along, and not actually reading them as a slower reader might).

  81. Here’s a Microsoft C program that scrambles the inner letters of words (but not too much):
    extern struct{char*a;int b;char*c;int d,e,f,g;char*h;}_iob[]; void main(){int a,b,d,e,f;char
    c,g[256];for(f=fgetc(&_iob[srand(time(0)),0]);f!=-1;)
    {for(;f!=-1&&!isalpha(f);f=fgetc(&_iob[fputc(f,_iob+1),0]));if(f!=-1)
    {for(e=0;(e<255&&isalpha(f))||(g[e]=’\0′);f=fgetc(&_iob[g[e++]=(char)f,0]));
    if(a=(e>3))do for(d=b=e-2;–b;rand()&1||(g[g[b]=g[c=g[a=0,b],d],d]=c),–d);while(a);
    fprintf(_iob+1,”%s”,g);}}return;}

  82. Wveoher siuteds tihs is vrey beord, but it is vrey aismnug.

  83. However in the case of “Stana” it could work out as Santa or Satan, and could become a case for words being also contextually understood.
    Sean wrote:
    It’d be pretty easy to discern within
    “Heer cmeos Stana Calus, hree cemos Staan Clsau, rgtih dwno Stnaa Cusua lnae…”
    but perhaps things would would be more often misunderstood when written as, “Mnay chrliend wpshpori Saatn on hghi hdlioays.”
    I’ll notify the PTA.
    “Stnaa” and “Staan” would be read santa and satan, respectivly, because of the last letter, either “a” or “n”. determining the word depends heavily on the first AND last letter.
    this is such a great topic. i would enjoy studying it further, or atleast reading someone eles research into this.

  84. arg, need edit button.
    first 2 lines should read:
    Sean wrote:
    However in the case of “Stana” it could work out as Santa or Satan, and could become a case for words being also contextually understood.

  85. The wohle tinhg gtes me werid, eevn thuogh it is vrey werid.

  86. Thanks, Justin! I really like this:
    “Artichoking to a reindeerwash at Cartilage Unfamiliarity, it dentis’t molar in wheat oater the ladybugs in a weed ate, the ozzy insideout thong is twit the foist and list louvre be at the ranft pekoe. The raft can be a Tyrol MIDI and you can shell rind it whippet pogrom. Toes is buncombe the harem mood digs not road energy ladder by insectivore, but the wild as a Wookiee.”
    Here‘s the direct link.

  87. Only today I saw this in english, and I must say it was really harder for me to read than in spanish (I have a good english, speak and read fluently). Anyway, here is the version from my email, for what it’s worth:
    Sgeun un etsduio de una uivenrsdiad ignlsea, no
    ipmotra el odren en el que las ltears etsan ersciats, la uicna csoa ipormtnate es que la pmrirea y la utlima ltera esten ecsritas en la psiocion cocrrtea. El rsteo peuden estar ttaolmntee mal y aun pordas lerelo sin pobrleams. Etso es pquore no lemeos cada ltera por si msima.
    La paalbra es un tdoo. Pesornamelnte me preace icrneilbe… SLAUODS Anrdés.
    So, the thing is that being spanish a language with a little of inflection, it’s not harder for a spanish speaker to read it than for you to read english.
    By the way, I found a response, wich I’m not gong to translate, but hopefully somebody will, about other two ‘studies’:
    Cuando recibí por primera vez el mensaje respondí:
    Un expesimento sinilar de una univarsibad divamapquesa denueftra qua las palebzas puenen teher una o das lejras epuivoxadas sin qne ezto inpkda la camprengión del mandaje. El extudeo grovocó uma enérqiha pratesto del Spndijato Dvnéz de Corractyres, qus ve pñligrcr su foente da trebejo.
    De inmediato, un amigo escribió:
    Hay o ro ex erime to se ejante de u a uni ersi ad a emana que mues ra que pu de se uir ley ndo e un t x o au cuan o des pare can al un s le ras de las pa ab as.
    Yau notr omash echoe nhun grias obr elau bicac ionde loses pacio sent rela spa labras.

  88. The order of letters in a word is important. What this example does not indicate is the context in which the words appear. Those of us who read this correctly were able to do so because we understood the context in which the words appeared from a grammatical standpoint. I bet you didn’t read it very quickly the first time, but because we were curious we went back and read it again, but probably faster. It may be true that the eye does not attend to every letter, but it’s also true that the eye does not attend to every word. In other words, we guess, a lot, and sometimes we guess wrong because of context. For example, I bet most of you would also be able to read this sentence and understand it if you do some guessing:
    Babe _______ hit the ______ over the __________.
    We can read this because we know the context in which it is written and we have the prior knowledge to fill in the missing words. But, what happens when we don’t have that prior knowledge or we don’t understand the context or grammatical structure. Obviously, we then attend to the word and all of its letters. That’s what phonics is trying to teach our kids to do. How many times have you heard a student read something and use a word that was similar but incorrect? If so, he’s probably not getting the context either. Have you ever been reading something and gotten to the bottom of the page and despite the sentence being continued on the next, you read the first word on the next page before you’ve actually turned the page? As a teacher, I’ve seen many students do it.
    By the way, it’s “Babe Ruth hit the ball over the fence.”

  89. I’m not a native speaker of French but I can read that French paragraph “snas porlblème”… should this make me really happy about the level of proficiency to which I have acquired the language?

  90. Mias oui!

  91. I sincerely hope this style of writing does not catch on. Yes it’s comprehensible but it also gives me a headache when I read it because I’m used to reading correct spelling.
    Also, why encourage everyone to spell wrong? Yes it’s the same end result (comprehension) but that doesn’t negate the fact that it is still wrong.
    It looks like a five year old wrote it. It looks terrible.

  92. Just to let people know, it is my research, it was my letter to New Scientist, and I have a paper copy of the PhD, but not very postable!
    Thanks for all your interest in my expeirments of 27 years ago. Keep up the good work!
    Graham

  93. Hey, thanks for dropping by, Graham! It’s nice to see the originator here. And for those who are interested, here‘s a link to a summary of Graham’s thesis, “The Significance of Letter Position in Word Recognition.”

  94. Mark (at October 1, 2003 01:52 PM) made a critical point that we don’t see or percieve each letter or even word. In one glance at a list of letters we “see” only 4 or five of them. But we can “see” more than 3 times as many when the letters form words. We anticipate and make logical guesses. The options we guess from are limited due to our own experiences with words and common syntax in the language.
    You may want to read “Reading-from behind the eyes” by Smith.

  95. My only question is: Were the subjects of this study “taught” to “read” by look-and-say methods? Having listened to such victims of modern pedagogy read aloud, I have concluded that they don’t look at the internal letters of a word anyway, only the first and last and then guess. Folks who were actually taught to read (that is, phonetically) do indeed slow down while deciphering this mess.

  96. So… you’re boasting about a pedagogical method that left you reading less efficiently? You’re proud of having a hard time reading scrambled text?
    *shakes head in bewilderment*
    At any rate, I was taught to read the traditional way, and I have no problem with the text.

  97. Hey, try this “Word Masher”: http://www.aurete.com/wordmasher
    Cheers,
    Gary

  98. Johnny Rocketfingers says:

    As it states on http://www.mrc-cbu.cam.ac.uk/~matt.davis/Cmabrigde/index.html ,
    “The sprehas had ponits and patles”
    Might come out as…
    The sherpas had pitons and plates.
    The shapers had points and pleats.
    The seraphs had pintos and petals.
    The sphaers had pinots and palets.
    The sphears had potins and peltas.

  99. Aramis Martinez says:

    about script kiddies/gamers: it’s called l33t, and it’s a gamer subculture thing (sort of, “if u c4n r34d/wr1t3 l33t, ur 1 of us”); read Megatokyo for more enlightenment about l33t. I understand l33t now exists in multiple languages.
    about the Spanish translation: any guesses as to why a 3rd-generation Mex-American who speaks little Spanish had a much easier time reading the garbled version in Spanish?

  100. Does anyone know who invented this theory?

  101. Ihad no problem reading the paragraph in english or spanish. I don’t read word by word, I read in groups of words………..

  102. Leanderthal says:

    This is so totally tight. I would like to know if there is an easier way to type these scrambled words. Also, please add more examples fpr us to practice with. Thank You

  103. Word scramblers:
    http://www.lerfjhax.com/funky.php
    http://www.aurete.com/cgi-bin/wordmasher.pl
    I also would like to say “fuckin’ rad.” I would suggest to anyone who is interested in this subject to read this entire page, as all the comments on it are insightful, interesting, and helpful, and also the discussion at http://nielsenhayden.com/makinglight/archives/003677.html .
    I found that I could read the original paragraph on ‘rdaenig’ easily, but I had real trouble with disemvowelled text, text in which random letters are replaced with spaces, and this:
    “Nreuuoms pmeeononnhs peossss uiapocmltecnd etaaoilxnpn; nwttdtsniinoahg, the pdseuo-snfiiiectc spssliiimm is not snfiiiectc and eieecndvs are oetfn mdanleiisg.”
    (Numerous phenomenons possess uncomplicated explanation; notwithstanding, the pseudo-scientific simplisism is not scientific and evidences are often misleading.)
    Until I saw ‘pseudo-scientific,’ I could not pick out a single word. Is this simply because the words are long, or because there are grammatical errors in the text? I think this is worth studying, but I have no background in linguistics.
    Indeed, the mind is a strange thing. I wonder if a day will ever come when a computer will be able to process things in this way. I think if someone does develop good linguistic AI, it will have to develop and mature like human brains do.
    And finally, I would like to say that Graham Rawlinson is a badass.
    -James

  104. Hpapy New Yaer!

  105. i’m doing a science fair project on this, i’m a freshman in high school. any help ya’ll can give me? since you all seem to know very much in the subject. i guess it is a psycological thing. the way you read it. or the way your brain functions. i’m not sure. i haven’t started the project yet, i’m a procrastinator. but i’m guessing the way with having the first and last letters stay in place but not the rest of it, is the quickest to read. what about any other methods? please help.

  106. I search for forum like this long time.You website is very good!I will come next time !

  107. Hello

  108. A post at Acephalous brought my attention to the fact that this thread had been besmirched by persons hawking wares not relevant to a language blog. My whim was to delete the URLs they were trying to link to while leaving the comments as samples of classic early-2004 spam style. I mean, how can I delete “I search for forum like this long time”?

  109. David Smith says:

    Being deaf, I read lots of mispelled words when people try to communicate with me. It has a lot to do with context. I can sometimes sense my brain substituting words for wildly mispelled words — changing letter order, changing letters, changing the length of the word. Interestingly, I found the early example with mixed caps and lower-case easiest, all caps almost as easy, and all lower-case the hardest. Might be practice, though. I work as a proof-reader and find that when I have to wade through Spanish, my mind does the same dance. Might be different for a native speaker, though.

  110. Just knocked up a quick scrambler to play with jumbling – it’s fasinating me more and more!
    Indeed, some words seem harder to recognise than others. Is is familiarity, or is it that jumbling letter-pairs such as ‘th’ freak the brain a bit.
    Anyway the url is http://the-hum.co.uk/try.aspx Theres no instructions, its really just for languagage hat folk. I may start to add ‘rule’ settings, like ‘Keep familiar letter combinations together’ etc…
    Mat

  111. Wow. It’s cool to see the whole thing unfolding real time, lo these many years ago.

  112. Glinda Creekbaum says:

    Throughout the great pattern of things you’ll secure an A with regard to effort. Exactly where you confused me personally was in the particulars. You know, people say, details make or break the argument.. And that could not be more correct right here. Having said that, allow me reveal to you exactly what did give good results. Your writing can be really powerful and this is possibly the reason why I am making the effort in order to comment. I do not make it a regular habit of doing that. 2nd, although I can certainly see the jumps in reason you come up with, I am not really certain of how you appear to connect the details which in turn help to make your final result. For right now I will, no doubt yield to your position however trust in the future you actually connect the facts much better.
    [I deleted the spam URL but I can't let the name "Glinda Creekbaum" vanish. Spam on, Glinda, spam on! -LH]

  113. Throughout the great pattern of things you’ll secure an A with regard to effort.
    In the grand design of things you get a B- for hard work.
    So which is it, Glinda/Joann/Loathesome, A or B-?

Speak Your Mind

*