February 2012 terminology revisions.
August 2011 Complete rewrite.
May 2011 original article
December 2012 I've started a blog on sentence spacing.

How many spaces at the end of a sentence? One or two?

The goal in spacing at the end of the sentence is to provide enough visual separation to make it clear to the reader that they have reached the end of the sentence, without adding so much space that it visually breaks the flow.

One thing is absolutely true: using the same sized spacing between both words and sentences (which I'll just call word spacing for the rest of this article) is vastly more popular today than in the past. Older texts mostly use extra space between sentences—spaces larger than those used between words (although there were always some that chose to use word spacing for sentences). Gradually during the 20th century, word spacing between sentences become more popular both in Europe and the U.S. with many (eventually most) professional publishers. This change is not in question. The questions this article addresses are: why did this change happen, and must we try to follow this trend?

It's a silly thing to get bent out of shape about how much space goes between sentences. And yet all over the internet we can see people doing just that. There are the so-called "monospacers", aggressively declaring victory and brow-beating the rest of us to follow their path. And then there's the two-spacers, singing "that's the way I was taught" all the way to their graves.

So this article is intended to sort out the mess, and put both sets of conformists in their places. For my part, I don't care how many spaces you put there. I just want you to understand that you are free to choose, and mock anyone who claims there is a "rule" that you are breaking.

Expert Opinion

For the record, I'm no expert on typography, or any other particularly relevant career except perhaps web technology. So permit me to take a quick look at what makes someone an expert. Surgeons are an example of people who are experts. Their expertise is in the human body, and in cutting and repairing it. This expertise was earned by years of training, and maintained by still more training. And the expertise is based on an extensive foundation of published research that is constantly growing and changing. This foundation is available for anyone to examine.

As I've read the various articles out there on this topic, I've seen it repeated quite often that typography experts have settled on one space between sentences as the only way to do things. And this or that typographer might be quoted. Although often people who are not typographers are also quoted. But what is the foundation of this bit of expertise? If they're really in total agreement, or at least vast majority agreement, I should be able to look this up. However there's nothing there to find. There is no basis for this bit of "expertise". (In fact, I can't even find evidence that typographers as a group are making this claim themselves.)

There are certainly research papers that have studied this issue from the point of view of "readability", which must certainly be the whole point, right? (Perhaps not.) But not only do these papers fail to come to a consistent conclusion, they fail to establish a consistent basis for comparison. One study finds that people read word-spaced text slightly more quickly than extra-spaced (and then hedges its own claims). However, is speed a good measure of "readability"? How about number of mistakes, or retention? Is there room in the equation for simple aesthetics?

With no real expert basis for choosing a spacing style, it's my opinion that this is after all just a matter of fashion, like the length of the skirt or wearing white after labor day. Some of the more quoted sources on the one-space rule include style guides, such as the Chicago Manual of Style. The word "style" should be a clue here. It's only a rule if you work for the University of Chicago Press. And their goal is to present a consistent style for all their published works. Another publisher certainly might choose differently.

The Point

So if there's no clear set of studies, no foundation for this allegedly expert conclusion on spacing, then why has the industry shifted so strongly towards word spacing for sentences? Why don't other publishers choose differently? The hint is the the word I just used, "industry", and while there are several factors at work, the most important answer I believe is the age-old answer to many questions we ask.


In our case the money manifests itself in terms of time and energy and general human laziness. Human costs. There are a few issues related to spaces and money. But the biggest one by far is brought about by the ambiguity of the period. Periods are not used only at the end of sentences—they are decimal points, and are used in abbreviations. So we can not blindly put two spaces after every period. We only want them at the ends of sentences. One might suggest that a sentence is a period followed by a capital, but someone like P.T. Barnum might beg to differ.

Typesetting Technology

From the time that movable type was invented all the way up to the end of the 19th century the printed page was created by laying out block letters one letter at a time by hand in the desired layout. This was the practice for more than 400 years. And in this time, extra spacing dominated. The professional typesetter was hardly bothered by the ambiguity of the period. Typesetting was slow work, and the typesetter was certainly aware of the flow of meaning and context in what he was typesetting. Adding additional space between sentences was hardly an extra burden.

But eventually, printing moved from a craft to an industry. With the introduction at the end of the the 19th century of hot metal typesetting, a keyboard could be used to set the type. At the same time the principles of the assembly line changed the way the world worked. The demands of modern business required progressively faster production. If a newspaper (or a book or magazine publisher) was trying to get out text as quickly as possible for the least cost, they needed to streamline the typesetting process. In this context, one space was better, because the typesetter didn't have to think about what the period meant. They didn't even have to read for context, they could simply copy the literal text into the typesetting machine. And a one-space rule left no opportunity to "mess up". The idea of visual appeal would never need to come into play in this decision. Using word spacing between sentences was already something that existed, and while not being the most popular at that time, it was also not considered wrong. Changing your style guide to word spacing would reduce errors, increase production, and reduce cost. There was no downside for a publishing company to move to word spacing, except an arguable loss of aesthetic appeal.

Once you accept this, you can see the futility in following style guides which were produced by the industry for use within the industry. Even submission guidelines which mostly favor word spacing are going to be linked to simplifying the work of the editor and publisher.

The World Wide Web

That's my hypothesis on the lost space, based on reading what lots of other people have had to say on the subject. I haven't seen direct evidence that actually shows that this is what happened. But it is a more sensible hypothesis than anything else that's been offered to this point. And I can move forward a bit into an area of my own expertise. I was one of the participants in the emerging World Wide Web standard in the early 1990s.

The Web uses HTML as its standard data format. All the versions of HTML up to (but not including) HTML5 are based on SGML (Standard Generalized Markup Language). The goal of SGML, its entire reason for existence, is to separate the content of text from the meta-information (in this case, layout information). SGML puts meta information inside "tags" which are blocks of text inside of angle brackets (e.g. <TAG>) The part that appears between the tags is content. Content doesn't care how much space there is between sentences, that's layout. So SGML naturally always collapses multiple spaces into a single space. After all, a space's only purpose is to separate words and other things from each other, and one space does that as well as twenty spaces. The only thing that more spaces can do would be layout, and that does not belong anywhere except in a tag.

So the space collapsing behaviour of the web comes from this, not from anything at all related to the aesthetics of the spacing. The goal of HTML at that time was to be a small simple language used to retrieve other document types via hypertext. No tag to mark sentences was ever introduced into HTML. I can't even find any record that such a tag was ever proposed. There was a discussion in 1993 on how to render spaces after periods, and the conclusion was that a single space made the most sense. Again, this had nothing to do with typesetting or readability or aesthetics, it was all about expediency:

Actually, I think most books don't do this any more. Newspapers certainly don't. It's way too much trouble escaping out all the abbreviations. It doesn't really add anything to readability.

I personally would prefer "one space fits all" as writers of HTML really shouldn't need to know the fineries of typography (What's next? Range kerning?), especially since it's not something 99.99% of people will notice (i.e. it won't even look "wrong").

-- Kenneth Chang, www-talk, July 1, 1993


The consensus seems to be that no one wants to bother with extra space after a period that ends a sentence. There is much confusion about why this was the traditional usage, but no matter.
-- Terry Allen, www-talk, July 2, 1993

So in the case of the web, we can be sure that the reasons for using a single space are technical, and not typographic or otherwise aesthetic. And also not to conform to any particular standard. This is what lead me to conclude that this was the likely motivation of the industry well before computerized layout existed.

Typewriter habits

One benefit of my reasoning is that it finally makes sense out of the typewriter nonsense. This was an issue that initially drew me into this subject—the inconsistency of the typewriter arguments that are offered. Many "monospacers" claim that two spaces on typewriters made sense because the letters are not proportional, and so you need the extra width "to even things out" or something like that.

This makes absolutely no sense. Essentially, the argument would be that typewritters with their oversized spaces (and oversized periods too) need to have even more space added between sentences, while proportional fonts which use little tiny spaces that are smaller than almost every letter do not need two spaces. This never made sense. It just became something that people would repeat simply because it had a familiar and comforting ring to it.

However, in light of the argument I'm making here, typewriters finally make sense. The typist, like the typesetter, isn't simply mass-producing text. Unlike the factory worker or the computer program, they will naturally pay attention to the context. And in paying attention to the context, they can choose to add additional space between sentences.

Another thing that may be at work in the two-spaces typewriter rule is that it largely came from teachers. Even if we don't agree that it improves readability, perhaps teachers found that it improves "gradeability"—papers were easier to grade if sentences were clearly set off from each other.

This practice has also had some interesting interactions with modern user interface design. Most word processors allow you to add extra space between sentences if desired. And modern software is smarter, and not so easily fooled by abbreviations. However our training in typing two spaces has had some interesting implications. Many word processors ignore the extra space you type. Some let you add space, but only if you keep trying to show that you mean it. And some software depends on waiting for two spaces as an aid to clarifying when you've reached the end of the sentence, sometimes going so far as adding a period if you type two spaces without one. This wide array of behaviour is most certainly another thing that influences submission guidelines, and their desire for consistent spacing in submissions. Again, we see the most likely explanation has nothing to do with aesthetics and readability, and everything to do with technology and convenience.


On my last revision of this document, I used the phrases "French spacing" and "English spacing" to refer to word spacing and extra spacing respectively. I realize now this has introduced some confusion. Some people only use "English spacing" to refer to the modern typewriter practice of using two spaces between sentences, and indeed it seems that this meaning may be accurate in terms of its historic usage. Furthermore, "French spacing" is not simply used to mean that one (word) space is used between sentences, but sometimes that (ironically) an extra space is also used before the punctionation, as at the end of this sentence ! I now belive that this very detail has contributed to the confusion: if the phrase "English spacing" was applied only to typewriter habits, then indeed "English spacing" was a new thing introduced along with the typewriter. I think some people would then interpret this to mean that extra spacing was also a new thing, although this is clearly not the case.

This underscores something else that is also true: using "two spaces" was definitely new with the invention of the typewriter. But this is of course because it would never have made since to count spaces before this—you'd just measure the space. I think this has contritubed to the typewriter receiving the blame for the use of extra space, when in reality it was just responsible for changing the way we talk about "space" versus "spaces".


So where does this leave us? Hopefully we can all just relax. If your boss or teacher demands two spaces then you can type two spaces without getting bent out of shape. If your publisher demands one space you can also please them with no loss of honor. And if you are writing for yourself, then you are free to follow your own aesthetic sensibility. You do not have to be oppressed by the conformists any more, no matter which side they are on. As for myself, I favor the more handcrafted look of extra space over the mechanized look of word spacing. But most of the time (for the web at least) I probably won't bother.

Do what looks right.

Addendum, Aug 2011. I've reformatted this web page to give you some control over the spacing between sentences. By default, it loads with roughly two spaces between sentences—one real space, and an extra 1/4 em (the part you can change). While there is no strict standard, one space is often about 1/4 em. An em is a unit of measure where the height of the font is always one em (at least this is a modern standard, the original meaning of an em is the width of the capital letter "M"). A typical web font might be 16 pixels high so one em equals 16 pixels, and if a space were 1/4 em it would be 4 pixels wide. The following table lets you add extra spacing (just click on a box), and shows you what the total spacing is with the very improper assumption that the spaces you are already seeing are 1/4 em. Note that back when typesetting was done by hand, typical spacing between sentences was 1/3 to 1/2 em, although sometimes even a full em or more was used (and sometimes word spacing was used).

≈ 1/4em (1 space)
+1/12 em
≈ 1/3 em
+1/8 em
≈ 3/8 em (1½ spaces)
+1/4 em
≈ 1/2 em (2 spaces)
+1/2 em
≈ 3/4 em (3 spaces)
+3/4 em
≈ 1 em (4 spaces)
+1px +2px +3px +4px +5px +6px +7px +8px

At smaller font sizes some of these spacings may not appear to be different. Also, I could have given you fuller control by removing spaces between sentences entirely, however that would harm the content since (for example) a cut-and-paste of text would have no space character at all where there should be at least one.

I've also written about different techniques for wider sentence spacing in HTML, including the technique I use here.

Some Links

A few of other articles in support of additional sentence spacing:
  • My Sentence Spacing blog
  • Why two spaces after a period isn’t wrong (or, the lies typographers tell about history) - Heraclitean River
  • Two Spaces After a Period | Ditchwalk

    My Home Professional Home TomSaraZac Home
    Email Work Email