The other day I received a mail that was a variation of something I had read many years ago on Slashdot, but it looked more like the chain letters some people like to torture me with. It went like this:

Only great minds can read this
This is weird, but interesting!
fi yuo cna raed tihs, yuo hvae a sgtrane mnid too
Cna yuo raed tihs? Olny 55 plepoe out of 100 can.
i cdnuolt blveiee taht I cluod aulaclty uesdnatnrd waht I was rdanieg. The phaonmneal pweor of the hmuan mnid, aoccdrnig to a rscheearch at Cmabrigde Uinervtisy, it dseno't mtaetr in waht oerdr the ltteres in a wrod are, the olny iproamtnt tihng is taht the frsit and lsat ltteer be in the rghit pclae. The rset can be a taotl mses and you can sitll raed it whotuit a pboerlm. Tihs is bcuseae the huamn mnid deos not raed ervey lteter by istlef, but the wrod as a wlohe. Azanmig huh? yaeh and I awlyas tghuhot slpeling was ipmorantt! if you can raed tihs forwrad it.

I went on trying to find the original article, and I did, but incidentally I stumbled upon a page that claimed that it was a fake. Combining that with the obvious change in style made me get interest into the matter. After some time finding out references going back in time, aided with the invaluable WayBack Machine service I could more or less reconstruct the history of this long-lived meme.

Apparently all started when a guy named Graham Rawlinson wrote his Ph.D. thesis for the Nottingham University in 1976 with the title "The Significance of Letter Position in Word Recognition". In this work he makes some experiments that show this phenomenon. A summary apparently made by the original author can be found here. The IEEEXplore magazine printed a verbatim copy of this summary in their January 2007 issue, the PDF is available for subscribers in this non-clickable address (I don't want to link to subscriber-only sites):

It seems that the thesis went undetected (by Google standards, that is) for some time, until an article that discussed how people could understand sentences even if the words were played backwards from a tape was published in the Nature journal. The article can be read in Nature's website under the title Cognitive restoration of reversed speech, and a comment in the same issue: Backwards Bohemian Rhapsody.

Apparently it was also published in New Scientist (I cannot find any content on the web), and then Rawlinson sent a letter linking these results with his own thesis from the seventies. A thread shorty thereafter appeared on Usenet.

I don't know what happened after this, but somehow it kept roaming in teh intarweb until 2003 when it gained critical mass. James F. Bisso blogged about it in May, 1st. In his post, the premise is to keep the first and last two letters instead of just one; and also proposes replacing all vowels with hyphens. After September, this entry will receive many comments, including Rawlinson's.

In September 12th, 2003, David Harris posted in a slightly modified version of the then-circulating text to experiment with meme dissemination, later explained in another post. The meme was picked up by many blogs in a few days: Languagehat, James Bisso posted about it again, and then again, and again. Even a Perl program to scramble text was published. But the fire became unstoppable after it was seen on Slashdot.

After that, everybody was talking about this for a few days, and there was even college assignments. Instead of talking about "an English university", Cambridge and other institutions appeared on the texts. Matt Davis, member of the Cognition and Brain Sciences Unit, in Cambridge, wrote a huge compilation of sample scrambled texts in different languages, including comments based on actual reading research.

Somehow, the meme survived for many years, and came back to me six years later.