vocabulary and Bible reading levels
We have had a number of BBB posts on reading levels of various English Bibles. Typically, reading levels are computed by algorithms that consider sentence length, words per sentence, and number of syllables per word. A number of us have commented over the years that reading level is more complex than can be determined by just mechanistic factors such as these.
One factor which is never included, as far as I know, has to do with vocabulary level, including the social register of a word. A word may compute at a low reading level according to an algorithm, but it may actually require a high reading level. For instance, the word "ascribe" is relatively short. It has only two syllables. It would contribute to a low reading level calculation. But "ascribe" is a word which is not well known to many English speakers today. Oh, sure, many of you reading this post understand and perhaps even regularly use this word, but many visitors to this blog are capable of operating well at high reading levels, due to levels of education and practice reading literature which has high level words.
Furthermore, reading level algorithms do not currently calculate for whether or not a word is currently in use or is obsolete. The word "thence" is mono-syllabic (a high level word itself!). It would calculate at a low reading level according to the typical algorithms. But many readers today, especially children in the typical grade levels for which reading levels are aimed, 3rd to 12th grade, are not familiar with the word "thence". So even if a computer algorithm says that a sentence with the word "thence" computes to a low reading level, the word "thence" would create a reading level problem for many.
I have been pushing hard these days to check the TNIV in order to submit revision suggestions to its translation team (CBT) before its annual January 1 submission deadline. (We still need checkers for several Old Testament books. See the green background poll in the margin of this blog.)
Here are some individual words in the TNIV I have spotted which seem to me should raise the reading level of the TNIV, even though the TNIV computes with the usual algorithms to between a 7th and 8th grade reading level:
As you think about this post, please do not forget that a Bible version is, whether intended or not, best suited to a particular audience, such as those who have completed a high school education, or those who are familiar with church language or theological terms.
We are not suggesting in this post that reading levels or vocabulary levels be "dumbed down". It is appropriate to use higher reading levels and words such as those noted in the list from the TNIV for more highly educated audiences.
One factor which is never included, as far as I know, has to do with vocabulary level, including the social register of a word. A word may compute at a low reading level according to an algorithm, but it may actually require a high reading level. For instance, the word "ascribe" is relatively short. It has only two syllables. It would contribute to a low reading level calculation. But "ascribe" is a word which is not well known to many English speakers today. Oh, sure, many of you reading this post understand and perhaps even regularly use this word, but many visitors to this blog are capable of operating well at high reading levels, due to levels of education and practice reading literature which has high level words.
Furthermore, reading level algorithms do not currently calculate for whether or not a word is currently in use or is obsolete. The word "thence" is mono-syllabic (a high level word itself!). It would calculate at a low reading level according to the typical algorithms. But many readers today, especially children in the typical grade levels for which reading levels are aimed, 3rd to 12th grade, are not familiar with the word "thence". So even if a computer algorithm says that a sentence with the word "thence" computes to a low reading level, the word "thence" would create a reading level problem for many.
I have been pushing hard these days to check the TNIV in order to submit revision suggestions to its translation team (CBT) before its annual January 1 submission deadline. (We still need checkers for several Old Testament books. See the green background poll in the margin of this blog.)
Here are some individual words in the TNIV I have spotted which seem to me should raise the reading level of the TNIV, even though the TNIV computes with the usual algorithms to between a 7th and 8th grade reading level:
abodeWhat are other words you have noted in the NIV or TNIV (or other versions) which seem to you to be above the 7th grade reading level?
abounding
acclaim
afar
amid
ascribe
avenged
bestowed
bloodguilt
encamp
endued
exult
fowler
gaiety
grievous
hallowed
highborn
ibis
insolence
lapis lazuli
lowborn
mallow
portent
rampart
renown
resound
respite
resplendent
unwary
vexed
As you think about this post, please do not forget that a Bible version is, whether intended or not, best suited to a particular audience, such as those who have completed a high school education, or those who are familiar with church language or theological terms.
We are not suggesting in this post that reading levels or vocabulary levels be "dumbed down". It is appropriate to use higher reading levels and words such as those noted in the list from the TNIV for more highly educated audiences.
posted by Wayne Leman at 8:09 AM
5 Comments:
I have always been a fan of the NIV. It was the first Bible I was introduced to. Then I started hearing about the Bible Experience and the TNIV. I was a little wary of it at first but as far as listening to the language in TNIV, I have enjoyed it thoroughly. I think it's definitely a lot easier for me to understand than the KJV.
Wayne,
Those of us working with adult learners of English use various word lists (developed by corpus linguistic methods). John Bauman, for instance, has a good set of links to three such lists under Vocabulary Resources on his website.
One of the easiest tools to use is Tom Cobb's (and Paul Nation's) VocabProfile, which takes your words input and classifies them by list. Using the English (not French) Profile, I input your 31 TNIV words. Only "gaiety" showed on any of the lists as included in the top 2000 most frequent words in the English corpus. The others were "off list" meaning not very common words at all.
Putting in John 1 TNIV, however, shows a completely different list. Of the more than 1000 words, 4 words (0.38%) are on the Academic Word List, 21 words (1.99%) are on the top 2000 word list, 916 words (86.82%) are among the top 1000 most frequently used words in English, and the remaining 100 or so words (around 10%) are not frequently used English words.
In contrast, John 1 KJV has the following breakdown: 5 words (or .47%) on the Academic Word List; 28 (or 2.65%) words on the K2 list; 833 (78.96%) on the K1 list; and nearly 190 words (or 17.91%), such as "shineth" and "begotten" and " verily" on any frequency list.
In contrast to John 1 (in either TNIV or KJV) is your post here. You wrote 528 words, 28 (or 5.3%) of which are on the AWL; 4.73% is on the 2K list; 76.52% is on the 1K list; and 13.45% are not frequent.
The other cool thing about this is you can get Anglo-Saxon percentages and Greco/Latin/Fr cognate percentages. I'll spare everybody (or let you do them yourself) as I'm not sure this says as much about readability of the words.
In theory, the more frequent the words (i.e., the ones on the 1K list first; then the ones on the 2K list; then the ones on the AWL) the easier the words to learn and to read.
For some of the research behind all this, there's a bibliography here.
Kurk, fascinating! I got an error message when trying to use VocabProfile with the Firefox browser, but it worked fine with I.E.
I'll have to do more with this tool. I think I may even have a list somewhere on my computer of all the words in the TNIV. It would be interesting to run that through these tools. Then it would be interesting to run all the words of other versions, such as the ESV.
I could start playing if I kept up with this stuff!! :-)
So, first off, I've found your blog to be one of the only reputable sources for intelligent discussions regarding hebrew/greek and translation...props!
So, I'm currently a student in seminary, studying both hebrew and greek...and there were a couple of words on that list of which I have no idea of their meaning...given, there's no context, but still...
I think our English language is changing, in the sense that our vocabulary is generationally oriented - those in the older generations are familiar with many words which those in my generation have never heard of. Same is true of the generation that grew up with the internet. I'm wondering how many more words will fade into oblivion? Seems like there will be a great need for new English Bible translations down the road...
anonymousperson is on to something. Language constantly changes, and proving the need for new translations are the next 2 posts on this blog:
idiomatic Bible translation
and
Do we think in words?
Wayne:
IE, Firefox, and Safari all worked for me to run VocabProfile. Haven't seen any "known issues" documentation so if you really want to use Firefox with the tool, then you might email Tom Cobb.
One caution: VocabProfile (at least the most recent online version for English) does not catch 2-word and 3-word verbs such as the one Iyov noted (i.e. "carried out" ) in his comment on your next post (and the others your mention there). If you do email Tom, ask him to add the sensitivity to such idioms. My guess is their research does include the idiom frequency. And the VocabProfile tool can handle it with a few code lines--already they handle contractions such as "did't" making ii "did" and "not" before processing.
Post a Comment
Subscribe to Post Comments [Atom]
<< Home