18.456 Wordcount revisited

               Humanist Discussion Group, Vol. 18, No. 456.
       Centre for Computing in the Humanities, King's College London
                   www.kcl.ac.uk/humanities/cch/humanist/
                        www.princeton.edu/humanist/
                     Submit to: humanist_at_princeton.edu

         Date: Mon, 03 Jan 2005 06:29:34 +0000
         From: lachance_at_origin.chass.utoronto.ca (Francois Lachance)
         Subject: Wordcount revisited or goibniu redux

Willard,

Wordcount whose existence you announced to the subscribers of Humanist in
posting 18.107 has proved for me one of the more interesting toys of 2004.
I too like many of those who reported their play experiences to Humanist
(see the Goibniu thread) hankered for an interface that provided
connection to a dictionary or at least to the corpus occurences in situ. I
want to know what some of the words (in some cases acronyms) meant. I
wanted a context of use. Frustrated, I chalked up Wordcount as a nice
implementation that unfortunately I could not access via a text-only
browser.

It was however in imagining an implementation of Wordcount for the
visually impaired that I came to a better understanding of how to read
frequency lists. Or at least one way of reading a large list. This
imagining was influenced by my experience with TACT. I played with
Wordcount to generate KWIC-like offerings. It was fun to look up names [of
frequent contributors to Humanist] and read the the words that occurred
before and after. The strings certainly provided surreal effects but more
importantly it was a step to reading the frequency list with less of a
focus upon the individual occurrence and more upon groupings. Memories of
TACT also led me to consider the joys of ranking.

Below is a ranking of gerunds related to perception and cognition. The
first number indicates positing in the Wordcount listing. Concerned as I
have been over the years with embodiment and mental faculties, I was
curious about the relative ranking of "thinking" and "feeling". The
relative ranking of "seeing," "hearing" and "touching" paints a vivid
picture of the traditional hierarchy of the senses that emphases the
distance of the "distance" senses from those that work in closer proximity
to the objects of perception.

It is also interesting to compare the ranking with the ranking generated
by examining the queries submitted to Wordcount. "Feeling" requested
almost twice as frequently as "thinking". "Touching" not yet or not
recently.

Listing from Wordcount and Querycount (as conducted Dec 30,2004)

Thinking 730 4412
Feeling 753 2265
Seeing 1590 7593
Hearing 1917 18010
Listening 2220 15357
Processing 2689 8601
Computing 4241 17364
Assessing 5554 \\
Touching 5595 \\
Viewing 6301 12025
Modelling 6637 19072
Calculating 9552 19636
Observing 8152 \\
Contemplating 11739 20620
Sensing 11790 23086
Detecting 13264 \\
Pondering 21862 \\
Analyzing 35384 \\

Cognition 24468 18227

Neither "cogitate" nor "cogitating" are in the 86800 most frequent words
found in the British National Corpus. Of course, this leads one to perhaps
query a grammatical set such as "feel, felt, feeling" or "see, seen,
sight".

Revisiting the relations between lexia, words, semantic fields: it seems
language is hardwired for hypertextuality.

I believe at some point in 2003, Willard, you posted to Humanist some
musings about interface and habitat and pointed to a 1997 article by Terry
Winograd in which it was suggested that computer use was moving along a
trajectory from computation to communication. Could it be that
implementations such as Wordcount are contributing to bringing attention
to the space between computation and communication that is the space of
comparison?

Between forge and crucible is the smith... many thanks to Norman Hinton
for bringing anvil, tongs and hammer to Goibniu. Not likely I would have
returned to Wordcount without that work.

-- Francois Lachance, Scholar-at-large http://www.chass.utoronto.ca/~lachance

A calendar is like a map. And just as maps have insets, calendars in the
21st century might have 'moments' expressed in flat local time fanning out
into "great circles" expressed in earth revolution time.
Received on Mon Jan 03 2005 - 01:54:08 EST