In a previous post, I explored the history of spacing after sentences, where we saw that the common practice for centuries was to include a much wider space after a period (or other mark that ended a sentence).  Since a double space nowadays imitates that practice—which comes from the era where the forms of many of our modern fonts were created—a double space should at least be considered an acceptable choice when typing or typesetting text.  Others will prefer the single space, but it is merely a preference.  I submit that there is room for both, and there is actually room for better typography in general, which could return to a more detailed distinction between different types of spaces with different widths.

I also summarized the history of the Chicago Manual of Style, which demonstrates the shift in preferences and values from 1900–1950, leading up to their current judgmental position where they only include information on sentence spacing to condemn anyone who would try to make such a distinction.  In a Q&A post, one editor explains this position and offers a number of rationales in favor of it:

[I]ntroducing two spaces after the period causes problems: (1) it is inefficient, requiring an extra keystroke for every sentence; (2) even if a program is set to automatically put an extra space after a period, such automation is never foolproof; (3) there is no proof that an extra space actually improves readability—as your comment suggests, it’s probably just a matter of familiarity (Who knows? perhaps it’s actually more efficient to read with less regard for sentences as individual units of thought—many centuries ago, for example in ancient Greece, there were no spaces even between words, and no punctuation); (4) two spaces are harder to control for than one in electronic documents (I find that the earmark of a document that imposes a two-space rule is a smattering of instances of both three spaces and one space after a period, and two spaces in the middle of sentences); and (5) two spaces can cause problems with line breaks in certain programs.

Let’s take each of these objections in turn.

  • it is inefficient, requiring an extra keystroke for every sentence

Yes, so is the space we put after the period itself.  Arguably, a period and a capital letter are plenty to signal the beginning of a sentence.  There were times in history when it was perfectly acceptable to write punctuation marks without any space after them, as Nicholas Jenson did when he printed the first ever Roman typeface (as opposed to earlier Gothic scripts) in the 1400s.  Many ancient documents that were written or chiseled into stone did not even bother with a space even between words.  (Youhavetoadmitthatyoucouldpuzzlethissentenceoutwithoutmuchtroubleifyouhadto.)  We have a lot of spaces, not to mention other typographical niceties, that aren’t “efficient.”  Why bother hitting the SHIFT key at the beginning of each sentence or proper name?  And while we’re at it, why not start lopping off silent letters from words?  Why not adopt “text-speak,” and type U for you, R for are, etc.?

Efficiency is rarely a consideration in the conventions of formal language usage, and condemning this one feature for inefficiency is arbitrarily singling one thing out for no apparent reason.

  • even if a program is set to automatically put an extra space after a period, such automation is never foolproof

That’s why we have copy editors.  In an e-mail or informal document, it doesn’t matter if one sentence happens to have the extra space missing.  The world will not end.  But in published matter, you have a copy editor to check for such things, just as that person should notice if a sentence begins with a lower-case letter, etc.  All of these arbitrary style conventions, down to making such proper punctuation is used consistently in every bibliographic citation, require some effort.  We used to hire people to put letters in one at a time by hand, and now we’re annoyed that we might have to check for the rare instance where our spacing isn’t automated?!

Besides, the automation can make it pretty easy.  Make the double-space the default for periods, put in a macro to check for common abbreviations followed by periods, and check the rest.  If a double-space is the default, it will stand out more if misused in the middle of a sentence, making it easier for a copy editor to spot.

  • there is no proof that an extra space actually improves readability—as your comment suggests, it’s probably just a matter of familiarity (Who knows? perhaps it’s actually more efficient to read with less regard for sentences as individual units of thought—many centuries ago, for example in ancient Greece, there were no spaces even between words, and no punctuation)

Well, that’s interesting.  The editor just disqualified her first objection with this argument.  As I already stated, spacing conventions are arbitrary, and some traditions were much more “efficient” than ours.  Why stop with eliminating that one space?  Why retain a space at all?

It’s true that studies have been inconclusive about readability.  But the point is that they have been inconclusive, which indicates that if there is an effect, it’s not significant enough to argue for double-spacing or for single-spacing.  Double-spacing may not increase readability under some made-up experimental conditions (and it does seem that there were flaws in the few studies that have been done), but it doesn’t decrease readability either.  So, all other things being equal, this isn’t an argument against double-spacing at all—it just says that we don’t have strong evidence for forcing a move to double-spacing . . . or to single-spacing, for that matter.

  • two spaces are harder to control for than one in electronic documents (I find that the earmark of a document that imposes a two-space rule is a smattering of instances of both three spaces and one space after a period, and two spaces in the middle of sentences)

Um, hire a decent copy editor.  (See answer to the second argument above.)  I bet your document that imposes a one-space rule also has instances of two spaces sometimes and occasionally two spaces (or even carriage returns, or a tab or two) in the middle of sentences.  So what?  You’re going to need to strip those errors out anyway.  And once you do, this problem reduces to the objection raised in the second argument above.  The solution, again, is hire a competent person to read through the manuscript and spot errors.  You need someone to do this anyway to check for all sorts of other trivial stylistic nuances.  A decent copy editor can also check for spacing; fire the ones who can’t.

  • two spaces can cause problems with line breaks in certain programs

Are you kidding?  Seriously: fix the programs.  This used to be a problem back in the era when machine typesetting started to become more common, but before manuscripts were commonly prepared on computers (i.e., the 1960s or so).  Those antiquated devices were incapable of dealing with small issues like two spaces, and so they would introduce line breaks between the spaces, causing a space at the beginning of the next line (which looked bad).

Nowadays, there is no excuse for this sort of nonsense.  If your publishing software is stupid enough to start a new line with an empty space, get a new piece of software.  This is not a hard problem, and the idea that some idiot didn’t program in a basic line-break check should cause us to modify our stylistic conventions, well, I don’t even know what to say.  If some bug in some publishing software didn’t support double quotation marks, should we abandon ever using a double quotation mark in prose?

What’s really behind these arguments

In the end, what can we really conclude about the single-space recommendation and the discussion used to back it up?  Essentially, it’s about laziness.  A double space is “inefficient”?  Well, I say a single space is lazy.  Double spaces are harder to automate and to make consistent in documents?  Well, stop being so lazy and hire a qualified copy editor, and stop depending completely on automation that is bound to mess things up.  If you can check for punctuation, capital letters, and even the formatting of bibliography, you can check for appropriate spacing.  In fact, no automation of spacing is ever perfect, even with only single spaces—if you don’t have someone checking that anyway, your publication is going to look terrible according to any typographer’s standard.

Yes, in the end, there is no proof that double-spacing is easier to read or “better” in some absolute way.  But there rarely is such proof for any subtle aspect of typography.  All we can say is that centuries of typographers thought the larger spaces after periods were necessary for syntax and, quite frankly, looked better.  A half century of other typographers disagree.  There’s no absolute answer that will be correct for all time in such a case.

But what we can say, as I discussed in my previous post, is that the single-space standard arose not because of any of the reasons mentioned by this Chicago editor, or because of its typographical beauty, which was apparently suddenly realized after centuries of following other conventions.  Instead, it was due to trends in publishing and a move toward cheaper and, frankly, lazier typography (which admittedly was also influenced by the easier to use, but problematic, technology).  It’s hard to put different types of spaces after a period or after various other kinds of punctuation.  We complain about the problems of automating a script to double-space, but can’t be bothered to proofread for spacing.

I’m ashamed of publishers today.  And any decent typographer should be as well, for forcing this lazy standard upon everyone.

I do believe that there might be some aesthetic value to having all spaces be exactly equal.  All of these supposed justifications are simply made up to excuse that aesthetic choice, but none of them offer any objective “proof” that the single space is better.  It is simply different.  And arguments can certainly be made the other way: historical priority, the importance of distinguishing a period following an abbreviation or initial from a period ending a sentence, etc.

If the Chicago Manual wants to declare a standard, let them do so.  But apparently a lot of people object to this single-spaced standard, unlike most issues of typography.  Perhaps there is a reason: a lot of people want that wider space.  And, considering it was the standard practice for centuries, is it really that important to the editors of the Chicago Manual of Style that we crush that tradition and bully everyone into some other arbitrary standard?

Can’t we just say it’s up to the designer, as any reasonable spacing issue should be?  And, for private correspondence or internal documents that aren’t typeset in a professional way, can’t we just leave people alone?