© 1993 Ken Glasziou
© 1993 ANZURA, Australia & New Zealand Urantia Association
Ken Glasziou, Maleny, Queensland
A previous study on this subject was published in the March/April 1992 issue of Six-O-Six. The study was undertaken because of speculation that The URANTIA Book may have been written by a single author. Some that have been suggested are: Dr. W. Sadler, Lena Sadler, Bill Sadler, Wilfred Kellogg, Carl Jung, H.G. Wells, and Robert Millikin.
Many forms of style analysis have been used in attempts to identify authors of various documents and literary works. The advent of computers has enabled investigators to use increasingly sophisticated methods to collect and analyze data. Much of the modern work has its roots in methods pioneered by Mosteller and associates. Included among these procedures is the fre. quency of use by individual authors of ‘marker’ words that begin sentences or join clauses and phrases within sen tences.
In the previous investigation, the frequency of use of marker words at the commencement of sentences was used in a broad-based investigation. For example, the first analysis simply compared Part 4 with the remainder of the book, it being expected that there could be differences in style between the authors for Part 4 (earthbound midwayers) and the authors of remainder of the book (celestial beings). A global test using six marker words indicated multiple authorship at a probability level of one part in more than 30,000 . The results of a survey indicated that there are not less than four authors of The URANTIA Papers, and perhaps nine or many more.
My collaborator in this work has been my son Paul, a medical graduate with a Ph.D. in maths and statistics, who suggested that it would be possible to apply these same techniques to compare writing styles for single authors who are named as the writers of multiple papers in the book. In this way, variability in the use of marker words in different papers by the same author would be taken into account when attempting to distinguish individual authors.
The papers selected for study were: Papers 11, 12, 13, and 14 by a Perfector of Wisdom; Papers 73,74,75, and 76 by Solonia; Papers 78,79,80, and 81 by an Archangel of Nebadon; Papers 82, 83, 84, 113, and 114 by the Chief of Seraphim on Urantia; and Papers 85,86,87,88, and 89 by a Brilliant Evening Star of Nebadon.
A number of marker words used to commence sentences were surveyed including ‘and’, ‘but’, ‘this’, ‘when’, ‘a’ or ‘an’, ‘it’, ‘as’, ‘many’, ‘by’, ‘thus’, ‘there’, ‘the’, ‘on’, ‘after’, ‘that’, and ‘neither’. The words ‘and’ and ‘the’ were chosen for the main survey on the grounds of highest frequency of use by all authors. The data for this survey are shown in Figure 1. A global likelihood ratio test in a logistical regression analysis showed that for the word ‘and’, the results indicated four or more authors at a probability level of .0003 . For the word ‘the’, three or more authors were indicated at a probability level of .0001 .
Individual comparisons were made for pairs of authors using pooled data for all the papers by each author and applying the Chi squared test for significance. The words ‘and’ and ‘the’ formed the basis of this analysis, but were supplemented by other marker words to further distinguish between authors. The results are shown in Table 1, and indicate the authors to be grouped together for the results shown in Figure 1, as well as drawing clear-cut distinctions between all of the nominated authors.
| Brilliant Evening Star | Solonia | Chief of Seraphim | Archangel of Nebadon | |
|---|---|---|---|---|
| Solonia | And, P = .025; The, P = .001 A or An, P = .001 |
|||
| Chief of Seraphim | And, P = .05; The, P = .001 But, P = .005; A or An, P = .001 |
And, P = .001; The, P = .005 A or An, P = .025; But, P = .05 |
||
| Archangel of Nebadon | And, P = .025; The, P = .005 But, P = .001 |
And, P = .001; The, P = .001 A or An, P = .005; But, P = .05 |
And, P = .005; But, P = .025 This, P = .005 |
|
| Perfector of Wisdom | And, P = .05; The, P = .005 But, P = .001; A or An, P = .001 |
And, P = .001; A or An, P = .05 As, P = .025 |
The, P = .005; But, P = .01 Many, P = .001 |
Table 1. The probabilities that pairs of authors of URANTIA Book Papers are different individuals as indicated by the frequency of sentences commencing with the marker words, “And” or “The” (see Figure 1.). When necessary, other marker words were used to obtain confirmatory evidence. All the authors are distinguishable from one another.
Together, these results indicate that each of the five sets of papers from The URANTIA Book were written by five different individuals as nominated in the book itself.
Reference
Mosteller F. and D.L. Wallace “Applied Bayesian and Classical Inference. The Case of the Federalist Papers” (Springer Verlag, N.Y., 1984)