Saturday, February 26, 2011

...When "most scholars agree..." at an appalling rate!

One of the most annoying and inaccurate statements that regularly appear in textual criticism articles is the infamous "most scholars agree..", usually activated when an argument or claim would otherwise appear weak and probably suspect.

Until now however, we haven't had a quantitative way of measuring the relative frequency of this phrase, and therefore gauge its true value and semantic content.

Well thats all over now, with "Google Ngram Viewer", which lets one search and chart the relative frequency of any phrase, searching the literature spanning over 100 years, which has been digitized or scanned.

First we searched the following related phrases:

(1) "Most scholars agree..." 
(2) "All scholars agree..."
(3) "proven beyond reasonable doubt..." (the control phrase).
click to enlarge

As expected, "ALL scholars agree" is less frequent than "MOST scholars agree...".

What is of course disturbing is that nonetheless, the phrase "MOST scholars agree..." has obviously shot through the roof, beginning oddly with the advent of Hort's Greek NT and the infamous Revised Version of 1882.

But someone might object, "yes, but there are more publications now than ever before."   But this is a false objection.  Our control-phrase, "proven beyond reasonable doubt" is a phrase expected to increase dramatically over the last few decades, because of the obvious increase in both court-cases and the publicizing of same (media frenzy), along with human rights issues, and the multiplication of legal arguments, cases, and lawyers writing on this subject and using this phrase.   But, the relative frequency of this word has only gone up moderately, as is to be expected, even though the volume of legal writing is vastly larger than writings about textual criticism.

Someone might say, yes, but there could be a divergence in both the quantity and rate of change between legal documents and textual critical articles.  So we ran the test again using another textual critical phrase, expected to increase with the same frequency:

(1) "Most scholars agree..."
(2) "oldest and best manuscripts"




We can be sure now, that "oldest and best manuscripts" is a great test-gauge of the actual expected increase of such scholarly phrases, because it has increased in usage an immense amount, due to its appearance on just about every second page of every "modern version" of the NT published since 1950.    This phrase has gone from the obscure backwaters of textual criticism of the 19th century to the lips of just about every Christian in North America who reads the Bible, one of the largest "captive book audiences" known.   Surprisingly, the phrase actually goes down in the literature, showing that scholars and popular writers are quite hesitant to use it, in spite of knowing it very well.

At the same time, it shows quite clearly that our real phrase of interest, "Most scholars agree..." has been multiplied exponentially and beyond belief, spiraling completely out of control in what can only be described as a hysterical fashion.

Is this just bad writing?   Or an 'easy way' for modern authors to artificially increase the strength of their polemic and apologetic claims?   The use of this phrase can hardly be being used unknowingly or unconsciously.  Its a standard weapon in propaganda and debate.   What this really shows is an appalling collapse in honesty in scholarly writing, and a ridiculous increase in absurd and inappropriate debating technique.

As a final control-test, we have plotted two synonymous phrases side by side, with near-identical results:

(1) "most scholars agree..." 
(2) "most experts agree..."



The result is predictable:  the two phrases are used and abused interchangeably, with about the same appalling exponential increase in frequency.   So much for 'scientific method' showing an overall improvement in accuracy and honesty.

mr.scrivener

1 comment:

  1. Thanks for posting this. As a philosopher "most philosophers agree" is a little tablet of poison that destroys the paper in which it appears.

    Thankfully, Chalmers quantified some stuff. http://philpapers.org/rec/BOUWDP

    This kind of survey (updated yearly) should be a regular requirement in social sciences.

    ReplyDelete