Page 1 : Beale Ciphers Analyses
BEALE CODES -- WERE THEY A HOAX?
This site is interesting for four reasons.
First, it presents a compelling argument that the three letters addressed to Robert Morriss, which, according to the Beale Papers, were written in 1822, were in fact written 20 or more years later. This argument is based on the evolution of english language vocabulary. Several words which were non-existent in 1822, but common many years later, were used repeatedly in these letters.
Second, it is equally convincing, by a comparison of the word usage and sentence structure, that the Beale Papers and the three Morriss letters were written by the same person.
Third, it purports to deliver a "knock-out proof" that it's all a hoax based on an interesting fact: The C2-DOI solution is valid only if the DOI text from the Beale Papers is used, and this text contains many errors !
Fourth, there are several pages dedicated to statistical analyses of the three ciphers. The main scatter plots are reproduced below.
The author refrains from reaching any conclusions, stating the following instead:
"I was uncertain whether the computer analysis ... had been explained clearly enough to be understood by someone without some background in the statistics of information theory. I finally decided in June 1988 to make this chapter public as it stands, with the idea that if anyone wants to understand it, it will be necessary to learn some statistics."
So, let's go to school. The following explanation of "Coefficient of correlation" comes from www.pinkmonkey.com Through the coefficient of correlation, we can measure the degree or extent of the correlation between two variables.
Perfect correlation: If two variables change in the same direction and in the same proportion, the correlation between the two is perfect positive. The coefficient of correlation in this case is +1.
Absence of correlation: If two series of two variables exhibit no relations between them or change in variable does not lead to a change in the other variable, then there is no correlation or absurd correlation between the two variables. The coefficient of correlation is 0.
Limited degrees of correlation: If two variables are not perfectly correlated or is there a perfect absence of correlation, then we term the correlation as Limited correlation. It may be positive, negative or zero but lies with the limits ± 1.
|Absence of correlation||Zero||
|Perfect correlation||+ 1||
|High degree||+ 0.75 to + 1||
- 0.75 to -1
In all three charts, the plot lines with round points, the "pure random", range from -1 to +1, showing an absence of correlation, as intended by the author for comparison purposes.
With reference to the plot lines using triangular points:
C1 ranges from -1 to +1, the same as pure random, showing an absence of correlation.
C2 is almost vertical between +0.75 and +1, showing almost perfect positive correlation.
C3 is a rapidly rising line between +0.5 and +1, showing a high degree of positive correlation.
The conclusion of this site is that the Beale Papers are a hoax because:
1. The Morriss letters and the rest of the text were written by the same person.
2. The Morriss letters were written 20 or more years later than claimed.
3. The C2-DOI solution only works if the erroneous DOI text of the Beale Papers is used.
4. The statistical analysis shows that C2 is distinctly different from the other two. C1 probably consists of random numbers, while C3 may, like C2, have a solution.
It is difficult to disagree with any of these ideas, all of which are disappointing to treasure hunters. But nagging questions remain.
What if the hoax was intended, not to sell pamphlets by James Ward, but to conceal and misdirect attention from true solutions to C1 and C3 ?
And in cryptography, isn't the main objective that the ciphertext appear to be random? Isn't it then possible that the coefficient of correlation for C1 and C3 is witness to their cryptographical success?
The "random" conclusion for C1 stands in stark contrast to the opinion presented in Page 3, which shows some very unrandom strings. There is, it seems, a statistical argument for every point of view.
Back to homepage