Question:
What are wordprints? What do they have to do with the Book of Mormon?
Answer:
As John Hilton put the matter, if wordprinting is a valid
technique, then this analysis suggests that it is "statistically
indefensible" to claim that Joseph, Oliver, or Solomon Spaulding wrote the
30,000 words in the Book of Mormon attributed to Nephi and Alma.[1]
The Book of Mormon also contains work written by more than one author. Critics
who wish to reject Joseph's account of the Book of Mormon's production must
therefore identify multiple authors for the text, and then explain how Joseph
acquired it and managed to pass it off as his own.
Neal A Maxwell
Institute
Wayne A Larsen and Alvin C Rencher, “Who
Wrote the Book of Mormon? An Analysis of Wordprint, Book of Mormon Authorship, (1982): “Our approach is sometimes
referred to as the science of stylometry, which can be defined loosely as
statistical analysis of style. It is also called computational stylistics. We
do not use the word style in the literary sense of subjective impressions
characterizing an author's mode of expression. We must deal with countable
items which are amenable to statistical analysis. We look then for what is
frequent but largely unnoticed, the quick little choices that confront an
author in nearly every sentence. Such choices become habits, so the small
details flow virtually without conscious effort.”
Detailed
Questions and Answers
What is a
wordprint?
Wordprinting, or
"stylometry" as it is more commonly known, is the science of measuring
literary style. The main assumption underlying stylometry is that an author has
aspects of literary style that may be unconsciously used, and can be used to
identify their work. Stylometrists analyze literature using statistics, math
formulas and artificial intelligence to determine the "style" of an
author's writing.
Because authors may write
on a variety of topics, the vocabulary they use may vary considerably.
Researchers often attempt to use "non-contextual words" in their
analyses to avoid this problem: patterns in the use of these words (e.g. such
as: and, if, the, etc.) will be less influenced by a
change in subject matter.
Debate about the value of
wordprints persists, though it has been used in some academic settings to
identify previously-unknown authors. Readers are cautioned that the results of
wordprint analysis of the Book of Mormon are only as reliable as they would be
for other written works, and that "the jury is still out" as to
whether wordprints can actually do what their advocates hope. The statistical
analyses are not generally disputed; the points of contention revolve around
the assumptions which undergird the statistics.[2]
Initial efforts
The initial Book of Mormon
wordprint studies were carried out by Larsen, Rencher, and Layton.[3]
They compared twenty-four Book of Mormon authors (each having at least 1,000
words) to each other, and concluded on the basis of three separate statistical
tests that these authors were distinct from each other and Oliver Cowdery,
Joseph Smith, Jr., and Solomon Spaulding.
These efforts were
critiqued in Ernest H. Taves, Trouble Enough: Joseph Smith and the Book of
Mormon (Buffalo, N.Y.: Prometheus Books, 1984), 225–60. John Hilton
characterized Taves' review as "fundamentally flawed," and noted that
his effort "therefore did nothing to add to or detract from their
work." [4]
An LDS author considered
some of Larsen, Rencher, and Layton's work in D. James Croft, "Book
of Mormon 'Wordprints' Reexamined," Sunstone no. (Issue
#6) (March-April 1981), 15–21. off-site
Croft pointed out some flaws in their assumptions, and was cautious about
whether wordprint evidence should be accepted or rejected as it then stood.
John Hilton and
the Berkeley Group
Methods
A more sophisticated
approach was taken by John Hilton and non-LDS colleagues at Berkeley.[5]
The "Berkeley Group's" method relied on non-contextual word patterns,
rather than just individual words. This more conservative method was designed
from the ground up, and required works of at least 5,000 words.
The Berkeley Group first
used a variety of control tests with non-disputed authors (e.g. works by Mark
Twain, and translated works from German) in an effort to:
·
demonstrate the
persistence of wordprints despite an author's effort to write as a different
'character'
·
demonstrate that
wordprints were not obliterated by translation (e.g. two different authors rendered
by the same translator would still have different wordprints).
The Berkeley Group's
methods have since passed peer review, and were used to identify previously
unknown writings written by Thomas Hobbes.[6]
The Berkeley Group
compared Book of Mormon texts written by Nephi and Alma with themselves, with
each other, and with work by Joseph, Oliver, and Solomon Spaulding. Each
comparison is assessed based upon the number of "rejections" provided
by the model. The greater the number of rejections, the greater the chance that
the two texts were not written by the same author. Tests with
non-disputed texts showed that two texts by the same author never scored more
than 6 rejections; thus, one cannot be certain if scores between 1–6 were
written by the same or different authors. Scores of 0 rejections makes it
statistically likely the two texts were written by the same author.
However, seven or more
rejections indicates that the texts were written by a different author with a
high degree of probability:[7]
# of Rejections
Certainty of being different authors
7 99.5%
8 99.9%
9 99.99%
10 99.997%
Results
The results are striking:[8]
Recall that any test over 6 indicates different authorship;
1–6 or less is indeterminate; 0 is same author. Each x represents one test.
Go to this link for charts and
endnotes. http://en.fairmormon.org/Book_of_Mormon_wordprint_studies
No comments:
Post a Comment