[torsoboy]

I just saw this tonight (although it has probably been around for years) and I thought it was pretty dang interesting.

http://www.echalk.co.uk/amusements/O...bledWords.html

I had no problem whatsoever reading it at a normal reading pace. Very cool/strange (I think).

And here's another one to try... our mind sure is a tricky thing! http://www.echalk.co.uk/amusements/O...HowManyFs.html

[Oisín]

You’re right, it has been around for years.

It’s also quite incorrect. It works for words up to about five, maybe six letters, because their visual representation can’t change too much by a few letters being switched around.

For words with more than around six letters, it doesn’t work—it requires that you only jumble groups of letters, not jumble the letters indiscriminately.

Example: from the text from that site:

Accdnriog to a raeecsrh at Cbdgmraie Ueiinrstvy, it d’nsoet mteatr in waht oedrr the lrtetes in a wrod are. The olny iaomrptt tnhig is taht the fsrit and lsat ltteer be in the rhgit pacle.

The shorter words are still fairly easy to read, but the longer ones require quite a bit of unjumbling in your head to be legible.

[analogika]

This was posted in 2004, and it was timelined *then*.

http://forums.macnn.com/89/macnn-lou...ingsh-wtirnig/

From that thread:

Origionally posted on Slashdot.org:
Follow-up: Can You Raed Tihs? meal worms writes "A Slashdot article appearing last Monday, which reported on the claim that scrambled words are legible as long as first and last letters are in place, was circulated to the University of British Columbia's Linguistics department. An interesting counter-example resulted:

"Anidroccg to crad cniyrrag lcitsiugnis planoissefors at an uemannd, utisreviny in Bsitirh Cibmuloa, and crartnoy to the duoibus cmials of the ueticnd rcraeseh, a slpmie, macinahcel ioisrevnn of ianretnl cretcarahs araepps sneiciffut to csufnoe the eadyrevy oekoolnr."

As demonstrated, a simple inversion of the internal characters results in a text which is relatively hard to decipher."

[tooki]

Oh god, not this crap again...

[KeriVit]

This the first time I actually felt a timeline was totally and completely necessary. I think I first read this 10 years ago!

[Eriamjh]

teilnmie!

[tooki]

What bothers me about the letter-order thing isn't even that it's old, but rather that it's junk science. The example text has letters swapped far more restrictively than it claims. If you actually try the rule as stated, most texts become undecipherable.

[brassplayersrock²]

Originally Posted by tooki 

Oh god, not this crap again...

*sings* memories

[torsoboy]

Originally Posted by tooki 

What bothers me about the letter-order thing isn't even that it's old, but rather that it's junk science. The example text has letters swapped far more restrictively than it claims. If you actually try the rule as stated, most texts become undecipherable.

I wonder what the real rule is then. Why are some messed up words (like those in the example paragraph) so easy to read, while others that are messed up (like teilnmie) much harder. Our brain is doing something to help us, but obviously it isn't just the first and last letter that does it for us.

[Oisín]

1. They don’t mix up letters across individual word boundaries (i.e., in ‘timeline’, they keep the ‘timeline’, and then they only mix up ‘timeline’ and ‘timeline’ individually; they don’t mix letters from ‘time’ with letters from ‘line’).

2. In longer, non-compound words, they don’t mix up letters at random, but keep them together in smaller, semi-coherent groups.

If you take ‘appreciate’ and turn it into ‘aciapprete’, for example, it’s still fairly easily recognised, because you still have the visual identity of cia and ppre in addition to the initial and final letter. You basically only have to switch those two groups around to recreate the word the way it was meant to be, which is a menial task for our brain. If you completely mangle the image of the word, however, it becomes illegible, because there are no recognisable starting points to base an interpretation on. ‘Aterppaice’ (inversion of internal letters) is useless for almost-immediate recognition.

It’s not true what this junk science page (and its ilk) says, that the human eye doesn’t read through an entire word and sees it only as one big image of a word; or at least, it’s not true for about 99 per cent of humans. The human eye does see words as just blocks of images, without reading through each word individually, but when it comes to longer words, the eye sees the words compound blocks of images, not single images. So as long as you only switch some of the internal ‘building blocks’ of the compound images around, you’re fine. As soon as you break up the building blocks as well, the eye is lost. With compound words, we automatically identify individual lexemes, so any jumbling of internal letters has to happen without crossing lexeme boundaries too much; in non-compounds, image formation is more arbitrary, but often corresponds loosely to syllables.

Taking an extreme example, it would be completely ludicrous to suggest that any Finnish speaker would be able to recognise the following as any word at all:

Eeeeiyyyääääääääääöödhjjkklllllmmmnnpr sssttttttn

Anyone can see that it’s completely illegible. However, somewhere in there, well hidden, is the word epäjärjestelmällistyttämättömyydellänsäkä änköhän (well known to be the longest non-compound Finnish word in existence). However, if you limit yourself to lexeme-internal jumbling, it becomes legible (if you’re Finnish).

Edit: For some reason, vB won’t accept words that long (sorry, Finland; sorry, Greenland), but the extra spaces inside that word are not supposed to be there, they’re vB quirks.

[SirCastor]

Originally Posted by Eriamjh 

teilnmie!

Ironically, My brain did not properly fix this one, and I tried to read it as mis-spelled.

[tooki]

Originally Posted by torsoboy 

I wonder what the real rule is then. Why are some messed up words (like those in the example paragraph) so easy to read, while others that are messed up (like teilnmie) much harder. Our brain is doing something to help us, but obviously it isn't just the first and last letter that does it for us.

In the sample text, with few exceptions, no letter is too far from its original location, and many parts of words weren't changed at all. For example:

original: Cambridge University
sample: Cmabridge Uinervtisy

Look at what's unchanged in "Cambridge" (bolded).

And then look at "University". It's not randomized at all, it was scrambled only within small groups: u-ni-ver-sit-y. No letter moved outside its little group.

Now I'm gonna take a shot at actually randomizing:

Cgradibme Uesvtinriy

Not so clear, eh?

If you google the web, you'll find great refutals of this "theory".

Oisin and I are both studied linguists (he much more than I), and the whole thing just ignores one linguistic principle after another.

[Chuckit]

Originally Posted by Oisín 

Taking an extreme example, it would be completely ludicrous to suggest that any Finnish speaker would be able to recognise the following as any word at all:

Eeeeiyyyääääääääääöödhjjkklllllmmmnnpr sssttttttn

Anyone can see that it’s completely illegible.

Looks about like any other Finnish to me.