Humanist Discussion Group

Humanist Archives: Jan. 16, 2022, 7:54 a.m. Humanist 35.463 - cautions about digital studies of words

              Humanist Discussion Group, Vol. 35, No. 463.
        Department of Digital Humanities, University of Cologne
                      Hosted by DH-Cologne
                Submit to:

        Date: 2022-01-14 07:33:29+00:00
        From: Willard McCarty <>
        Subject: digital studies of words?

In Humanist 35.459 Henry Schaffer reported on a central result from a
statistical study of words concerning, roughly, rationality vs
intuition, from 1850-2019, indicating in the authors' words a shift to,
"irrelevance of factual truth in public discourse", or more cautiously:
"All in all, our results suggest that over the past decades, there has
been a marked shift in public interest from the collective to the
individual, and from rationality toward emotion."

Note their word "suggest". One is hardly surprised. Because it is easy 
to come to the conclusion indicated, then one must ask about the 
rhetorical force of having arrived at these results by computational 
means. Is this not just the first step in study which would go beyond 
isolated words? Of course there will always be careless readers and 
leapers to conclusions who take such 'suggestions' as proof. But I 
would think that more caution, more qualification and less readiness to 
publish is called for. Are such analyses the computationally powered
shortcuts to the forecourts of truth that they seem to be? Should we 
not take into consideration the appetite for proof-by-machine that 
no doubt could be suggested by the same means?

Full disclosure: I've done the same, though with less sophisticated 
statistical tools, and I also have used 'suggest', I fear with insufficient 
caution. Thus we warn ourselves.


Willard McCarty,
Professor emeritus, King's College London;
Editor, Interdisciplinary Science Reviews;  Humanist

Unsubscribe at:
List posts to:
List info and archives at at:
Listmember interface at:
Subscribe at: