Joi Ito's Web

Interesting observation.

When I do crosswords I see words as single objects. They drop in to the slots in the puzzle. Rarely do I find myself fussing about individual letters.

Inetretsing osebrvtaoin

That reminds me of every 15-year-old that has ever instant-messaged me.

Similar to Andy's point, the cause of this capability may be developmental. When we first learn to spell and make mistakes the patterns are similar.

As a follow-up you can see what some linguistics bloggers are saying about it.

Here
http://www.bisso.com/ujg_archives/000224.html
and here
http://www.languagehat.com/archives/000840.php

Amazing. I read half of it before I realized some words were misspelled.

The conclusion that we read whole words is without merit. The correct conclusion, which is nothing new, is that there is a lot of redundancy in language, and this trick focuses on an area of English redundancy. It won't work in a highly inflected language, for obvious reasons.

That is so cool... ;)

Change the first and last letters though, and I'd be totally lost.

Fcuk sckos

this is very much like something i've been working on for several years, and that is; collecting UnCodes... Text that appears to be in code, but isn't...
Thanx for another Method...!

you have to know the word "correctly" before you can "raed" it in any order.

I cd rd tt wh no pm wr!

One of my english teachers came into class one day and just began writing on the board as follows (or something close to it):

"Tdy w wll nt s ny vwls"

It was a rather cool way to start class discussing the purpose of vowels and such.

But the letters have to be in order, twmseoha. Otherwise the drwo bljuem on the puzzles page wouldn't be any fun.

This is simply a play on the fact that there is no valid spelling of any word:

http://www.tmcdonald.com/blog/_archives/2003/8/28/1857.html

you have to know the word's correct spelling to be able to read it in the revised format - if you don't know the word at all, you're out of luck. but, with traditional reading, the first-time reader can suss their way through unfimilair words.

Wow, I used to think that I might have some very occasional and mild form of dyslexia, a form of it that didn't really effect me much, but now I realize I'm pretty normal. Nice to know.

I wonder if this applies only to the English language? I would imagine that it wouldn't work in languages like Finnish or Estonian were you often have many repeat letters in words...

perl -pe 's{(?<=\W\w)(\w\w+)(?=\w\W)}{ my $s; $s= rand(100) > 50 ? qq|$s$_| : qq|$_$s| for (split //, $1); $s; }eg;' < inputfile > outputfile

You can change the 50 part to something else to control the shuffling; 0 is not shuffled at all, 100 makes internal letters be exactly backwards, 50 is somewhere in between.

Spanish:
Mi nobrme es Moanlo y tbajrao en una eprsmea de icroimátnfa.

I think there is a problem with long words, they are quite hard to understand. Short words in Spanish are easy as in English.

re: long words

perhaps we break long words up into subsections?
m.

Yes - it's readable...but highly unpleasant.

I have absolutely no trouble reading that paragraph at all. I read in large chunks of words though, or in entire lines on a page the width of those found in mass market paperbacks. Which is how I can justify re-reading Cujo for the seventh time; it only takes me three hours.

Fascinating stuff, either way. I never would've figured it'd be that easy to read crap. Or, like the earlier comment by Andy, maybe we've all just been instant messaging too much.

What do they say about the signature line from the Dicky Roberts movie "Nucking Futs"? ;)

Amazing! However, I assume this applies primarily to native English speakers (readers). I wonder how well this applies when the language at issue is a second or third language?

English is not my first language and I had no problem to read it but like Manolo says in Spanish could be quite different.

It works, but only for words that are easily recognized or for sentences where the meaning is clear. Otherwise, plays with context and subtle changes in spelling which change the whole meaning are lost.

It seems to me possible that this has something to do with perceptual continuation/grouping.

http://www-psicologia.psibo.unibo.it/bib/92.htm

http://amazon.ece.utexas.edu/~qasim/papers/iqbal_pris01/node4.html

http://www.perceptionweb.com/perabs/p30/p3082.html

I saw a Norwegian variant on this a few weeks ago; I'm a native anglophone and my Norwegian is still pretty broken, though I read better than I speak better than I hear. The Norwegian variant kept the first and last *two* letters of each word (but then Norwegian jams several words together a lot of the time, so words are typically longer). I had no more trouble reading it than reading Norwegian normally; in particular, I was able to work out what it said despite not knowing many of the words.

Years ago I discovered that I can read an English text if I can only firmly identify one word in about three (letters from family members with idiosyncratic hand-writing). I think this jumbled-letter deal is similar; the human brain does very sophisticated delocalised pattern-matching as a matter of course.

The flip-side of our ability to read garbled text is that proof reading is hard, precisely because we see what was meant and don't see what's actually there on the page.

Just goes to show what tricks your mind can play on you, sometimes even without you realising what is happenning :-)

You should fix the title, even if it does break the links. There is no excuse for bad grammar.

is there a proper link? or is this just a casual observation.

Your mistake in the title also has a 'psycho-' linguistic explanation! Lots of research has been done on what that kind of error tells you about the human language processor; they sometimes call the phenomenon 'attraction' because it's like the agreement on the verb is 'attracted' to the word that came right before!

Your mistake in the title also has a 'psycho-' linguistic explanation! Lots of research has been done on what that kind of error tells you about the human language processor; they sometimes call the phenomenon 'attraction' because it's like the agreement on the verb is 'attracted' to the word that came right before!

It's eaceillonptxy aidnnostug how many sdeloppsuy kabdeegollnwe scceiimnooooc choorts will slrup up uabeinrtttud pcdeeiiinossstuc maaeklry.

Order doesn't matter, huh? "English University", huh? WOW, well if someone you probably don't even know tells you that someone told them that it happened at an English University, then it must be true!!

Dollars to donuts it's an urban myth and will show up on snopes.com soon.

Granted, the topic is very interesting. But don't take the story hook line and sinker.

-Truth

I wonder how this relates to touch typing.

This begs the question - how many words are there in the English language that have the same letters but with the middle ones in a different order: e.g. "From" and "Form" and does it make a difference?

This may apply to the 'natural' process of learning a language, as a native speaker. When a language is learned otherwise, aspecially from the book (by foreign speakers), the process is different and more prone to logic than 'instinct'.

Well....taht's geart nwes! You suohld tell my tecaher to sotp tankig mraks off my eassys for splleing msitakes!!!!

[Qoute]Comment from Truth on September 16, 2003 03:12 AM

It's eaceillonptxy aidnnostug how many sdeloppsuy kabdeegollnwe scceiimnooooc choorts will slrup up uabeinrtttud pcdeeiiinossstuc maaeklry.

Order doesn't matter, huh? "English University", huh? WOW, well if someone you probably don't even know tells you that someone told them that it happened at an English University, then it must be true!!

Dollars to donuts it's an urban myth and will show up on snopes.com soon.

Granted, the topic is very interesting. But don't take the story hook line and sinker.

-Truth
[/Quote]

I'm totally with ya!

-Annoyed-

Context is a major factor.

Knowing the language, and more particularly - the words being used, is the main factor.

Don't you have an extra "ch" in "rscheearch"?

pphaers tihs snecente is pfceert porof tihs is copmltee and uettr blthuslit... it had me gnoig trehe for a mnmeot tutohh...

Understanding the context is key!

From: Restoring the Vision, Government Technology Volume 16 Issue 12 Aug 3 p15 sidebar

A phenotropics presentation Jaron Lanier gave at the University of California at Berkeley included the following quote, for which no origin is found.

"“ ... randomising letters in the middle of words [has] little or no effect on the ability of skilled readers to understand the text. This is easy to denmtrasote. In a pubiltacion of New Scnieitst you could ramdinose all the letetrs, keipeng the first two and last two the same, and reibadailty would hadrly be aftcfeed. My ansaylis did not come to much beucase the thoery at the time was for shape and senqeuce retigcionon. Saberi's work sugsegts we may have some pofrweul palrlael prsooscers at work. The resaon for this is suerly that idnetiyfing coentnt by paarllel prseocsing speeds up regnicoiton. We only need the first and last two letetrs to spot chganes in meniang.” "

Lanier tied this quote to phenotropics by saying language theorists seem to make a mistake in assuming natural language is more like computer language than it probably is.

My phenotropic work examines a radically anti-parse/anti-protocol approach to computer science,” he said. “I naturally wonder if natural language can also be understood with less reference to the kinds of structure usually found somewhat in common between it and traditional computer science. What if comprehension could survive all manner of scrambling, particularly of word order and word variation choice, like tense? My hypothesis is that even a tiny amount of the mysterious element we call ‘context’ can correct even very high levels of scrambling.”

@yk: "This may apply to the 'natural' process of learning a language, as a native speaker. When a language is learned otherwise, aspecially from the book (by foreign speakers), the process is different and more prone to logic than 'instinct'."

Try it out with a non-native speaker. Like me :-)

Look again (at the small words)
"to-at-an-it-in-the-in-a-are the-and-at-the-can-be-a-and-you-can-it-a-is-do-not-by-but-the-as-a" are all spelled correctly. This suggests that context plays a role in our ability to infer meaning. (This was pointed out to me by linguist Sharon Klein.)

I think the title should have been Leettr oderr de'onst mtetar

My understanding was that we don't even read words one at a time but take in an entire gestalt including the line above and the line below and some words before and after our point of focus. I think we're always formulating the sentence we're reading as we go, anticipating likely next words and phrases.

It has already been said, but I'd like to confirm it: it works with english as a second language. And it works in french as well. But the words must be really scrambled inside, if you make the words with a logical pattern inside (grouping systematically the same letter and use natural ordering), you focus on the pattern and the effect is lost.

One strange thing, I've referrers from this page to my blog but there is no link at all and no reason to be linked as well... referrer abuse?

Um... I don't know about all this.

What it illustrates most clearly to me is the importance of the words of three letters or less as the connective tissue in almost any sentence of coherent English.

Here's a test, using only the words that are too small to scramble:

Now is the of our by son of

Think about it for long enough and you should be able to fill in the rest of this famous quotation without the remaining words - scrambled or otherwise.

Context (not content) is king.

[btw - if you really can't get it, I'll blog the answer over here: http://michaelocc.com]

ok, I tried the english and the german version - translated and scrambled by jimmiz (www.20six.de/jim). I am a native german speaker. Fact is: I had less problems with the english version than with the german version. I believe that is mostly due to the fact that the english words are shorter.

Xian is right; context definitely plays a huge part here. That takes care of all the words using the same beginning and ending letters, but having some of the middle letters switched, such as form and from, or peruse and persue.

The real killer here is long words, and I think this is why:

"perhaps we break long words up into subsections?"

You could safely jumble up the letters within the subsections of the words, and we'd still be okay, but if you jumble the whole thing it can become impossible, as demonstrated by "Truth". One fun word to play with is comprehension, and you can soon turn it incomprehensible: chineseomporn and cornishmepeon just for starters. :)

This issue has been talked about quite a bit. I think the fascination is that most people consistently spell words incorrectly and this is almost a way for them to say, "see, spelling doesn't matter". I suppose that spelling has to be correct at one point so that a learning process can take place.

Ajoconatne....

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
preo la paalbra es un tdoo.
Pesornamelnte me preace icrneilb

I got as far as the word English before I noticed anything wrong. Just proves the point. (although clearly there's some controversy concerning why this is the case; my money is on the whole-word theory). Y para manuel, puedo leer sus palabras también. Parece que es lo mismo en español.

I replaced your grammatically incorrect subject line with "The New Cognitive Hypothesis." Check with me for your future subject line needs.

"I wonder if this applies only to the English language? I would imagine that it wouldn't work in languages like Finnish or Estonian were you often have many repeat letters in words..."

I read the same text in Finnish yesterday and it was much harder to understand. I guess it would've been even more difficult if I hadn't read the original English text earlier.

Someone asked about if this only works for native English speakers and the answer is no. Me and all my Finnish friends were able to read through the text with no problems whatsoever.

A very interesting research result I must say...

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;(e3))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;}

as much as i want it not to make any sense, it does. weird!

Your title is just fine. Headlines have their own grammar. I remember reading some academic treatments of this in my college days, but the best I could find quickly on the web is here:

2.- Grammar. Newspaper headlines also have a special grammar, which is different from that of ordinary sentences. The main features of this grammar are: (a) omission of articles and the verb "to be"; (b) a special tense system; (c) the very frequent use of nouns as adjectives.

Does anyone knows who the first person was who wrote about this theory?

Also competing in "De nationale krakercompetitie"?

You can create your own texts more easily with this tool: http://www.wordmix.com

Suprisingly I was able read that paragraph without a problem though English is not my first language.

Any grammarians notice that the title of this page should be "Ordering of letters DOESN'T matter?"

The subject is the gerund "ordering," which takes a singular verb.

I think that your speed and accuracy of reading will affect your ability to read a "jarbled" sentence, let alone a paragraph. I have noticed that people who cannot read very well the "proper" spelling of things, often have trouble reading the "abnormal" version above. I must say though that on a side note, the people who created english language did not make a very well thought out plan for its grammar. And, yes I DO realize the horrible grammar I tend to use myself.

Great story, thanks. Best regards from Germany.

It makes it especially difficult to read when the words are spelled incorrectly. For instance:
It's eaceillonptxy aidnnostug how many sdeloppsuy kabdeegollnwe scceiimnooooc choorts will slrup up uabeinrtttud pcdeeiiinossstuc maaeklry. =>
It's exceptionally astounding how many supposedly knowledgeable socioeconomic cohorts will slurp up unattributed pseudoscientific malarkey.
As above, "pcdeeiiinossstuc" should be "pcdeeiiinossftuc"

I'm interested in a point that I haven't see brough up here; That is the claim that this was discovered during a study at an 'english university' or, 'cambridge' as I've also heard. I'm going to go with skeptical until proven true.

funny how ppl wanna blast this guy for posting something thats been shown and published and even after they make up their own jumbled sentences, that are still highly readable to anyone with an iq over 50, they still wanna flame the guy. i think some people need to turn off the pc and jog around the block to help with their depression.

Check out the following link. This dissertation was the probable source for the work.

http://www.mrc-cbu.cam.ac.uk/~matt.davis/Cmabrigde/rawlinson.html

Great story, reading skill has an effect because I has able to read even the sentances that suposedly renderer the theory useless. I think that the smarter you are the better you can read this format. English is my second language and I can read it fine. Also I tried this with Afrikaans but it did not work probalbly because of the tendancy of joining words into 30 letter superwords

Scott:

"pphaers tihs snecente is pfceert porof tihs is copmltee and uettr blthuslit... it had me gnoig trehe for a mnmeot tutohh..."

is it just me, or does this comletely contradict what he was trying to prove. this theory that we see words as a whole is skewed because our language has deteriorated to very short words, but you would have to be and idiot if you can't read the above.

That is very cool. I was wondering if there are other sites that has those kind of sentences with the words all scrambled. Email me at Godkindoflife@aol.com if you do know a site. Thank-you!

Actually, your title is indeed correct. Do not refers to "ordering" not, and since ordering is singular, don't is more appropriate to use than doesn't.

It is in the syllables, long compound or suffixed words we get confuse when reading because we read them as two words(ex. knowledgeable: knowledge and able). Also words that we personally are not familiar with or dont use we find it harder to figure out (ex. pseudoscientific)

tanhk you so mcuh! All of you just helped me with a science experiment! Yay for you!

is it just me, or does this comletely contradict what he was trying to prove. this theory that we see words as a whole is skewed because our language has deteriorated to very short words, but you would have to be and idiot if you can't read the above.

is it just me, or does this comletely contradict what he was trying to prove. this theory that we see words as a whole is skewed because our language has deteriorated to very short words, but you would have to be and idiot if you can't read the above.

I was trying to find this paragraph for a friend, while my nearly-five-year-old was sitting on my lap. When I found this site and highlighted the paragraph to copy / paste, my son gasped, "What does THAT say?" and started trying to sound it out. So I decided to record him trying, and then record him reading a corrected version.

I suppose my 'research' is skewed in that I did go ahead and tell him someone had just messed up when they were typing. Before I told him that, he'd given up on the first word, saying it was too hard. I didn't read it out loud to him at all, and some words were new for him, so he got some things wrong even in the second reading, but still I thought the way he handled it was interesting. I'd be curious to see how other new readers handle this paragraph. I'll post the mp3s on my site. -V

i lkie taht vrey mcuh but sotp rpeaitng the smae tinhg tiwce

Hi,

Could you please help me where or how can I find that research.

Thanks in advance.