Navigation:   TowerBells Home  => Unrelated Essays  => Sentence spacing

Sentence spacing —
a long version and a short version

This essay was prompted a TidBits article, which in turn had been prompted by an article from The Verge linked there (which you should also read) on this subject, based on a report that "Microsoft is weighing in on the debate, setting future versions of Microsoft Word to mark two spaces after a period as an error."  I wrote a long comment about that article, but upon reviewing comments already posted by other TidBits readers, I realized that much of this ground had already been covered.  So I transformed the draft of that comment into the long version that you see below, and posted to TidBits the short version that follows it.

The long version

Microsoft is wrong, and single-spacers are inconsiderate of their potential readers.  Here's why:

In the centuries when type was set by hand, before the invention of typewriters and Linotype, all fonts were variable-width.  So were the spaces that accompanied each font.  There were "en-spaces" which were the width of the letter "n" in that font, and "em-spaces" which were the width of the letter "m", plus thinner and fatter spaces of several kinds.  Typesetters were trained to use en-spaces as the standard spacing between words and em-spaces as the standard spacing between sentences, for readability.  Spaces of other widths had two purposes, both related to the fact that the full length of every line had to be identical in order to lock the type into the frame before it went into the printing press.  Either the ends of lines had to be padded out to that length with slugs and spaces, or additional space had to be inserted between words in order to justify the right ends of the lines.

When the typewriter was invented, with its fixed-width characters, double-spacing at the ends of sentences was adopted in order to emulate [ha!] the visual function of the em-space that was customary for printers.  But that did not change the methodology of typesetting used by printers, though typists simply abandoned the whole concept of right-justification of text.  Certainly nobody at that point abandoned the idea that wider spacing at the ends of sentences improves readability.  For one thing, that policy treats abbreviations as words, not as sentence endings.

It was laziness and cost savings that drove increasing abandonment of variable spacing throughout the middle of the 20th century.

The invention of HTML and Web browsers just [ha!] made the situation worse, because it explicitly adopted the convention that any amount of space was both logically and visually equivalent to a single space of some standard size.  Now it didn't make any difference if a nervous thumb bounced on the spacebar, or a lazy thumb laid on it — all was forgiven as a single space resulted.  As a result, the importance of variable spacing was forgotten, though for younger people it had never been learned in the first place.

Fortunately, the inventors of Unicode understood the conventions and methodology of traditional typesetting, and they made it possible to carry those conventions into the computer age by defining a wide variety of whitespace characters.  You can find a concise list with examples at emptycharacter.com, and an exhaustive treatment of the subject at the Wikipedia article on whitespace characters.  Both of those articles are, however, merely descriptive of usage, and do not specifically address the reasons for using various kinds of spaces within text.

For slow readers (e.g., those who read one word at a time), variable spacing might be relatively unimportant.  The same might be true for those who are happy with sentence fragments, either for reading or for writing.  But for fast readers (e.g., those who read a line at a time) and for those who care about how sentence structure conveys meaning, more space between sentences IS important.  And for those writers who do care about conveying meaning accurately — writing complete and grammatically correct sentences for the benefit of their future readers — sentence spacing is as important as any other element of writing.

The idea that proportional widths of characters make variable spacing unnecessary is wholly illogical.  As I have already shown, the idea that "With a typewriter, it makes sense to add an extra space to make clear that the sentence has ended" is totally false; that was NOT why the second space was added!  Besides, with a typewriter, the period that marks the end of a sentence is already wider than it would appear with a variable-pitch font, so it is therefore more visible.  If a double space is needed with the wide period of a typewriter, it is certainly needed with the narrow period of a variable-pitch font.

Q.E.D. — Microsoft is wrong, and they are encouraging their customers in laziness and/or thoughtlessness.

If you don't care about how I personally arrived at this conclusion, skip the next paragraph.

I learned to read very early, and have always been a voracious reader of everything that came into my field of view.  So I soaked up spelling and grammar and sentence structure from books and magazines that were well written and well edited.  In junior high school, I took a one-semester class in print shop, and learned some of the tricks of the trade of hand-setting type to print business cards, receipt forms and other small stuff.  In high school, I took a one-semester class in personal typing, and my parents bought me an office-grade typewriter for writing term papers, converting class notes into readable study notes, etc.  A scientific career brought me into contact with so many different kinds of computers that at one point I could touch-type on more than twenty different keyboard layouts.  God gave me a proofreader's eye for detail and a generalist's eye for the big picture, so for as long as I can remember, I have automatically spotted typographic errors while reading and I have been concerned with accurately expressing my ideas in writing.  When the World Wide Web came along, presenting an opportunity to publish information that I had been collecting for years, I wrestled with the single-space convention of HTML (mentioned above) until I found a satisfactory workaround.  So on my Website (www.TowerBells.org) you will find that almost all paragraphs of text have wider spacing at the ends of sentences.  And I will now give away the secret for how I accomplish that easily and gracefully — simply attach a non-breaking space (HTML entity name "nbsp") to every period that ends a sentence where another follows.  The word "gracefully" was used in the preceding sentence because there are other ways of making wider sentence-end spaces that are not so graceful, e.g., they might occasionally force unwanted space at the beginning of a wrapped line.  The text that you are reading has been composed with an editor that displays in fixed-pitch fonts, enabling me to double-space between sentences so that I can more effectively proofread what I have written.  I am quite aware that once I post this to TidBits, that sentence spacing will be lost; so be it.  That's a problem for another day.  (In converting both the draft comment and the short actual comment into HTML for this article, all double spaces were converted so that the visual effect would survive.)

This archived article explains variable spacing even better:
https://web.archive.org/web/20171217060354/http://www.heracliteanriver.com/?p=324

and it was followed by a very interesting comment from H.Geeves:

"The hilarious thing about it all is that this article is *not* really about whether a single space or double space is ‘better’ after periods. It’s about a different paradigm in typesetting in which the spacing after periods was apparently larger and more variable. But this fact *has* to be viewed alongside another fact alluded to in the Luckcombe graphic [--] that authors in olden days were less ‘sententious’; that is, they used longer sentences containing clauses, sub-clauses, and availing themselves fully of the semi-colon. The comma, the semi-colon, and the period (or full stop) in those days had a double function as logical markers *and* musical notation: the length of the pause was different between a comma and a semi-colon – and between a period and the rest. Thus, the spacing after the period should be viewed as part of the punctuation. After a long, clause-rich sentence, the pause provided by the period *and* its spacing allowed for a much-needed cognitive break. The greater liberty with post-period spacing was possible in part because of the fact that prose was not broken up into tiny soundbites for today’s drooling masses – thus no (or far fewer) pigeon-holes."
(This being a quotation in its entirety from another source, the inter-sentence spacing has NOT been converted from its apparently original form.)

In other words, variable-width spacing can serve as a visual cue to someone who is reading aloud, regarding audible spacing in speech.

The short version

I wrote a long comment, and then discovered that the "heracliteanriver.com" article cited by "David C." said essentially the same thing but with extensive research support.  So my long comment is being posted as an essay on my own Website (www.TowerBells.org), and I'll just make a short comment here.

The author of the article from "The Verge" referenced by Adam's note wrote, "Typewriters used monospaced fonts, ... so the extra space after the [period] was needed to make it more apparent that sentences had ended.  Word and many other similar apps make fonts proportional, so two spaces is no longer necessary."  That assertion is both factually false and logically false.  It is factually false because the first typists were merely imitating what skilled typesetters had done for centuries — use wider spacing at the ends of sentences.  It is logically false because if an extra space is "needed" following a period that takes up just as much width as every other character in a monospaced font, then it is even more needed after a period that takes the least possible width of any character in a proportional font.

If you want to write with single spaces everywhere, you are certainly free to do so.  And if you are an editor, it is certainly your prerogative to require all of your writers to follow your chosen style.  But to assert that double spacing at the end of sentences is somehow "wrong" is to insult the intelligence of everyone who chooses to respect this particular aspect of literary history and the principles involved in it.


This page was created on 2020/04/26.

Please send comments or questions about this page to the Webmaster/owner.
Use similar links on other pages to send specific comments or questions about those pages.