Book of Mormon Vocabulary

[Jamie123]

Two years ago I posted this:

The Book of Mormon Zipf Index - General Discussion - ThirdHour

Well my paper on this has finally been published in Computer Speech and Language - though for the past year it's mostly been lying on an editor's desk! You can access it in a link which I'd be happy to give you in PM (I don't want to dox myself on the open forum) but only for the next 50 days. Open access publishing is a little beyond my means, and this is not a prestigious enough journal for my employers to fork out.

BTW I now have a much better model for vocabulary growth which I'm going to publish soon, but I daresay it'll be another year before that finally gets past the reviewers.

The Book of Mormon does have an interesting vocabulary curve which you can see in Fig.6. The sudden spurt of new words about 40,000 words in seems to occur somewhere in 2 Nephi. Does anyone know the book well enough to suggest what might be happening here?

Edited May 11, 2021 by Jamie123

[Just_A_Guy]

1 hour ago, Jamie123 said:

Two years ago I posted this:

The Book of Mormon Zipf Index - General Discussion - ThirdHour

Well my paper on this has finally been published in Computer Speech and Language - though for the past year it's mostly been lying on an editor's desk! You can access it in a link which I'd be happy to give you in PM (I don't want to dox myself on the open forum) but only for the next 50 days. Open access publishing is a little beyond my means, and this is not a prestigious enough journal for my employers to fork out.

BTW I now have a much better model for vocabulary growth which I'm going to publish soon, but I daresay it'll be another year before that finally gets past the reviewers.

The Book of Mormon does have an interesting vocabulary curve which you can see in Fig.6. The sudden spurt of new words about 40,000 words in seems to occur somewhere in 2 Nephi. Does anyone know the book well enough to suggest what might be happening here?

Whereabouts, in 2 Nephi, is the 40,000th word?

You do see more extensive quoting of Isaiah in 2 Nephi (1 Nephi quoted extensively from what textual critics would call “Deutero-Isaiah”, but in 2 Nephi we see extensive quotation from textual critics would call “First Isaiah”).  You also see quotations of extensive sermons from Nephi’s father Lehi and brother Jacob, who (I believe, speaking anecdotally) have different voices than Nephi’s.  (Lehi at 2 Ne 1-3, Jacob at 2 Ne 6-10, Isaiah at 2 Ne 12-24).

As far as the production of the modern text goes:  the scholarly consensus is that Joseph Smith and Oliver Cowdery produced the first part of the current Book of Mormon (1 Nephi through Words of Mormon) last.  We know that early on in producing the text, Smith rotated through a wide variety of scribes; so one might expect to see a lot of new words popping up in the books of Mosiah and maybe Alma.  But Smith and Cowdery had been working together for some time when they finally got around to producing 2 Nephi; and if you assume that they are the actual authors (as opposed to translators/revelators) of the text—there seems to be little reason to expect that 2 Nephi would materially differ from their earlier work product.

Grant Hardy of the University of North Carolina-Asheville has written a book called “Understanding the Book of Mormon” that delves into these kinds of issues, which you may enjoy.

Edited May 11, 2021 by Just_A_Guy

[Vort]

Can you describe the meaning of a high Zipf vs. a low Zipf? A high Zipf would mean that the particular word collection you're looking at becomes less common, while a low Zipf would mean that the particular word collection is more common. So e.g. if relatively common words had a high Zipf, that would suggest that these common words occurred less often than you might expect. By the same token, e.g. if relatively uncommon words had a low Zipf, that would suggest that these uncommon words occurred more often than you might otherwise expect.

Do I have that right? What other implications or inferences might one reasonably draw?

I'm PMing you, btw.

[Jamie123]

Ummm....yes I think you are more-or-less correct. Under high Zipf index SOME high-frequency words would have much higher frequencies than other high-frequency words. I suppose the best way to put it is that if you rank words 1, 2, 3... from the most frequent to the least frequent, Zipf's index specifies the rate at which the frequency drops as the rank number increases. A low Zipf index means that the frequency drops slowly, meaning that words are closer to each other in frequency. High Zipf index means that the higher frequencies are higher and the lower frequencies are lower. The crazy thing is though that Zipf's index is not a constant even within the same document - generally it tends to be lower for the high frequency words than for the lower frequencies, meaning that the frequencies of commonly used words tend to be closer to each other than the less common words words. There is another version of the law developed by Benoit Mandelbrot (the fractal guy) that accounts for this somewhat - I find it not terribly accurate for high frequencies, but it sometimes provides a "fudge factor" to make the overall model work for the emergence of low-frequency words (which are what cause vocabulary to grow). Sorry if that's not very clear but it's getting late - LOL.

Edited May 11, 2021 by Jamie123