A science of typography? Comments on Wheildon's _Type & Layout_

[Stan Brown <stbrown -at- NACS -dot- NET>]

Thanks to Emily Skarzenski <eskarzenski -at- DTTUS -dot- COM> for posting her
summary of Colin Wheildon's book _Type & Layout_ (Strathmoor Press,
1995, ISBN 0-9624891-5-8). I was planning to produce my own summary,
having just read the book myself, but since we focused on different
points my remarks may still be of value.

Firstly, I think Wheildon made a valuable contribution: he measured
comprehension, as opposed to mere legibility. Studies of legibility
(the ability to distinguish one word from another, and the speed of
doing so), even numerical studies, are not new. But Wheildon studied
comprehension, the sine qua non of technical writing, and he did so
numerically, so that now we can say that a given typographical choice
will likely increase comprehension by 22%.

Nevertheless, I question how far Wheildon's work is applicable to tech
writing. Some comments I've seen on the TECHWR-L Digest claim, for
instance, that Wheildon proved that we should be setting technical
manuals justified rather than ragged right. In fact, his work does not
warrant such a conclusion.

Wheildon's studies were of advertising and of newspaper-length
articles. The examples in the book are all of this form, and the point
is hammered home again and again in the text. Indeed, he begins
Chapter One with, "Typography is the art of designing a communication
by using the printed word. It is employed in making newspapers,
magazines, books, handbills, posters, greeting and business cards,
pamphlets, and brochures ... anything that is meant to be read.
[ellipsis in original]" Note carefully the list that is in apposition
to "anything that is meant to be read": his focus clearly is not on
books, manuals of procedure, or scientific articles.

Since Wheildon did not study technical writing (at least, as far as we
know from _Type & Layout_ he did not), we must ask seriously whether
it is valid to extend his results to our area.

The types of writing that Wheildon studied differ from technical
writing in several ways:

  - They are much shorter, often only a few paragraphs long.
  - Their purpose is often to persuade rather than to inform.
  - They are read once only.
  - They are read from the beginning to the end, or at least sequentially.

By contrast, tech writing is quite often of book length; its purpose
is at least 95% informative; at least some parts of a given piece will
often be read many times; and readers typically jump into the middle
and expect to find what they need. This last point seems to me the
most important.

I think that we can learn from much of what Wheildon said. For
instance, Wheildon found that setting body text in a sans-serif face
degraded good reader comprehension by 82% (67% of readers had good
comprehension with Garamond, 12% with Helvetica: pages 22-23) and
increased the frequency of poor reader comprehension by 364% (14% poor
with Corona, 65% poor with Helvetica: page 59). In this case, it's
almost overwhelmingly plausible that material of greater length should
be set in a serif face for greater comprehension. But we must remain
wary of interpreting Wheildon's findings outside their proper sphere.

To take an extreme example, Wheildon's principle of Reading Gravity
(pages 37-51) is stated as a universal. But would it be valid in
Israel, where right-to-left reading is a significant tradition? or in
Japan, where reading is in columns? (I don't mean is it valid for
Hebrew and Japanese; I mean is it valid for English as read by persons
who have grown up with Hebrew or Japanese.) Even more important, to
what extent is Reading Gravity applicable when we expect our readers
to dip into the middle of a page, read a few sentences, and put the
book away? Though Reading Gravity may make sense as it is explained,
we cannot _necessarily_ assume that it applies to the writing we do.

Similar questions need to be asked about his other conclusions,
including his deprecation of ragged right text and headings in color.
It is true that we want our readers to comprehend what we write. But
they read in a different way from the readers of Wheildon's test
materials. Before we are in a position to adopt his rules, we must see
studies of technical writing along the same lines as his studies of
advertising.

We can learn much from Wheildon. But we must learn not just his
conclusions but his spirit of inquiry and his skepticism about what
"everyone knows."

Regards,
Stan Brown, Oak Road Systems      Cleveland, Ohio  USA      +1 216 371-0043
email: stbrown -at- nacs -dot- net                   Web: http://www.nacs.net/~stbrown