En dash = width of n?

James Puckett

Is there a British tradition of making the en dash the width of n? I am reviewing a British book and I don’t want to assume the author is wrong if this is just a British thing.

Thomas Phinney

No, the author is wrong.

For a long time, the en dash was a sort: it came from a standard set of extra bits that were shared across typefaces. It was half an em in width, a.k.a. half the current point size.

When it started getting built into individual fonts, that tradition mostly continued. But not everybody follows it. In particular, some type designers note that if the typeface is particularly condensed (or expanded) it may not make sense to keep the en and em dashes at their traditional widths.

James Puckett

I got in touch with the author who said he was just using it as a rough approximation.

Marisa Torres

So is the size of the en dash and the em dash left to the designer choice?

John Hudson

So is the size of the en dash and the em dash left to the designer choice?

No. The advanced with of the em dash should be equal to the body height of the type, and the advance width of the en dash should be half of that amount. If that isn't the case, then you're not making em and en dashes, you're making something else.

The trouble is that because em and en dashes were a common feature of metal typography, they got enshrined as such in text encoding, and there is still no encoding for 'design-specific proportional dashes'. So some designers hijack the em and en dash character codes for proportional dashes, but I think this is a bad idea.

Thanks for the reminder to discuss proportional dashes with some members of the UTC when at the Unicode conference next month.

Nick Shinn

I start out making my dashes the “proper” width – but it doesn’t always look good, so I often depart from convention.

There’s nothing wrong with that—it exploits the opportunity for enhanced functionality afforded by technological improvements that now include the dashes in standard font encodings, where they can be designed to be typeface-specific, rather than constrained by the requirements of being in a generic sort.

By the same token, one now includes italic figures in italic fonts!

There may also have been a benefit to em- and en- sized dashes in composing tabular matter, but that kind of “hard” spacing is now done by tabs.

***

Also bear in mind that the amount of the character width occupied by the glyph — in other words, the width of the sidebearings — is open to interpretation.

John Hudson

There’s nothing wrong with that

Except that what you've made are not em and en dashes.

I really believe that fonts should contain proportional dashes of, at least, narrow, medium and wide widths. But calling any of these em and en dashes when they are no actually em- or en-width isn't the solution.

PabloImpallari

The advanced with of the em dash should be equal to the body height of the type

John, by "body height", are you referring to the BBox Limits? UPM? or Ascender + Descenders sum?

Nick Shinn

John, you are proposing to shut the stable door after the horse has bolted.
What is to become of all the fonts with “proportional” em and en dashes?

Pablo: http://typophile.com/node/27742

Paul van der Laan

An em dash where the width is equal to the UPM is barely usable for current day typography in my opinion. Therefore we always tailor de width of en/em dashes to the horizontal proportions of the typeface, just as we do to the hyphen.

The width of the en dash is considered within the context of numbers, such as “1982–1988”. The width of the em dash is usually around 160% of that.

LeMo aka PatternMan aka Frank E Blokland

A nice thing of en and em dashes that respectively have a length of half and full UPM (and zero side bearings), is that one can simply make lines with these that have a perfect weight in comparison with the rest of the font. One can include (contextual) alternatives for other purposes IMHO.

Craig Eliason

That line effect can be achieved even with dashes with sidebearings, by adding emdash-to-emdash kerning to close the gap.

Paul van der Laan

Littering the underlying text stream with em dashes merely to create visual effects doesn’t sound very appealing to me.

George Thomas

In phototypesetting days, the Alphatype system and some others used set widths which were appropriate to the family. A 16-pt regular face in a family might run on a 16 set while its condensed regular face might run on a 12 set.

Since to a large extent in those days the set width for a family was dictated by the width of the widest character in the boldest weight, usually the /W, using such a system to determine a width for the /EM and /EN is quite easy and keeps it perfectly proportionate to the design of the face.

Marisa Torres

I just have seen a lot of uses of the em-dash, more in contemporary publications and design. This might become popular culture in time, or who knows maybe it will be forgotten. Would you recommend to create em-dashes properly and maybe add an extra glyph(alternate) for a longer or more proportioned "em-dash", so its more of a designer aesthetically choice? Or is this too much?

Martin Silvertant

Except that what you've made are not em and en dashes.

Historically that might be the case, but I feel you're taking the terms too literally. I've never measured my dashes but designed them optically. The em-dash particularly I find too obtrusive in many typefaces. I would either design a long, thin em-dash without much sidebearings or a shorter one with larger sidebearings and roughly the same thickness as the typeface. I don't know what those sizes would be, but I do know that sticking to fixed dimensions will result in dashes I consider either too long or too short. I thought optical adjustments were essential to modern type.

When it comes to setting prices ($75,–) I find the en-dash a bit too long, so I include an alternate en-dash which is in between the length of the hyphen and the en-dash and add an OT feature for it. Perhaps I'm the only one who finds the en-dash too long for this context, but to me this shows the importance of optical design rather than sticking to strict measurements and rasters and such. I feel this overly rational consideration of measurements is something from the 17th/18th century. It lead to some interesting designs but not optimal ones.

Since to a large extent in those days the set width for a family was dictated by the width of the widest character in the boldest weight, usually the /W, using such a system to determine a width for the /EM and /EN is quite easy and keeps it perfectly proportionate to the design of the face.

I think that's great to get a framework of the dimensions, but I can't imagine not having to do optical corrections, even if it's 1% difference.

Jan Schmoeger

overly rational consideration of measurements is something from the 17th/18th century

Sorry, hot metal, including hand-set type, was with us for two more centuries, 19th and 20th. This is where em- and en-rules and spaces really belong, imo.

Martin Silvertant

That wasn't quite my point. My point is that we have such advanced tools now that making optical adjustments is both easy to do and recommended. Conventions in measurements should be followed as guidelines. After that optical adjustments are made.

In my opinion one should not blindly follow rasters and defined measurements but judge these glyphs in context and adjust accordingly. Honestly I don't care what the length of an em-dash is supposed to be if it just doesn't look right. Do we want to conform to classical standards or do we want to make type that looks good? I'm not saying they're mutually exclusive, but the latter should be the bigger consideration.

John Hudson

Pablo:

John, by "body height", are you referring to the BBox Limits? UPM? or Ascender + Descenders sum?

UPM. That's the standard definition of body height, inherited from metal typesetting. In other words, one em is always equal to the nominal size of the type. An em at 12pt is 12pts wide. The only way to ensure that is the case is to make the width of em character glyphs equal to the UPM value of the font.

George Thomas

@John Hudson‌ -- What you say is the standard is correct, but in Univers 49, for example, would a full-width EM dash look good to you? It doesn't to me.

John Hudson

Martin:

Historically that might be the case, but I feel you're taking the terms too literally. I've never measured my dashes but designed them optically.

So you're not actually making em and en dashes.

I really, really do agree that what I call textual dashes, i.e. dashes as punctuation characters in text as distinct from dashes related to page-level structural elements, should be proportional to the typeface design. Ergo, there really should be proportional dashes provided for in Unicode (or, alternatively, a normative acknowledgement that the existing em and en dash characters might be proportional and a secondary set of 'strict' em and end dash characters encoded).

Kent Lew

(or, alternatively, a normative acknowledgement that the existing em and en dash characters might be proportional and a secondary set of 'strict' em and end dash characters encoded).

This would be the more practical strategy, I think, to satisfy most parties.

And, in fact, it seems to me that u+2014/u+2015 almost achieves this for em dash already — although not intended as strictly as you might wish, I imagine.

While conventions for the “em” dash as a parenthetical marker have drifted toward a more “textual” approach (and here, I should add the caveat that I am talking primarily about the American English practice, British convention preferring u+2013), those typographic traditions that use the “em” dash for a quotation dash have continued to prefer a more literal em, I believe.

In the Unicode Standard, u+2014 is acknowledged as the dash used parenthetically and u+2015 as a quotation dash. It seems to me these definitions could just be amended as you suggest — acknowledging that the nominal em dash u+2014 may in fact be proportional and that u+2105 should be a strict, literal em.

I think it would be more practical to steer bibliographic citation toward sequences of u+2015 (or some other new codepoint), for example, than to try to recover from many years now of using u+2014 parenthetically and expecting a proportional, textual dash.

Ben Blom

So is the size of the en dash and the em dash left to the designer choice?

Yes. Ultimately, the font designer decides what the glyphs of the font he designs, will look like. That’s his job.

A font designer should not be a historian, but a pragmatic tool maker. Choose, as a font designer, any tradition you want as a starting point for your design activities—and then create something that works for you (and for the future users of the font in question). The designer decides what mix of tradition and innovation to pursue. (Of course, drifting too far away from current traditions, will lead to a dysfunctional font. Drifting too far away from the current used sizes of the “en dash” and “em dash”, will lead to a dysfunctional “en dash” and “em dash”.)

By the same token, a font designer chooses the x height of the font he designs. However, when he chooses an x height which is too far out of step with the current traditional x heights, many users will consider that font to be dysfunctional.

So you’re not actually making em and en dashes.

The exact meaning of a word, can change over time—especially when the context in which this word is used, changes over time. If a lot of people call something an “em dash”, then it can be considered to be an “em dash”. Not calling a glyph an “em dash”, because it is 1% longer or 1% shorter than its traditional length, is just silly to me.

Russell McGorman

I think that, in as much as they are called em and en dashes because those are their names, no mater what the width, they'ed still be em and en dashes―And being digital instead of metal, there's no particular reason for a literal interpretation.

kupfers

Old-fashioned German typographer here. I prefer the hyphen short and on a 1/4 UPM or narrower (so it’s easy to ID and not confused with a dash), the en-dash long and on 1/2 UPM (because I use it with wordspaces or thin spaces, and as a minus for convenience or instead of cents in prices), the em-dash on 1/1 UPM but shorter than the em, leaving small built-in sidebearings (because I like to use it without wsp.). That is for normal width typefaces. I can see that compressed designs might call for different proportions. And I can see that you guys won’t agree on how to make your dashes. That’s okay (probably) but I changed the dashes of fonts to my liking more than a dozen times because, as a typographer, I want to decide how my typesetting looks like.

John Hudson

For those people who think that the relationship of names and the things to which they are applied don't matter: would you or would you not expect an em dash and am em space to be the same width?

George Thomas

...would you or would you not expect an em dash and am em space to be the same width?

In today's advanced technology, my personal preference is no for display, yes for text.

Nick Shinn

I would if I thought about it, but then I would think, Who uses the em space these days? Sure, it is in the “Insert White Space” menu in InDesign, but if one wants to indent a line by an amount equal to the type size, I suspect almost all typographers would use either a tab or a paragraph indent—because those are global commands that can be applied to multiple paragraphs, and they also scale when text boxes are resized.

What are the other contemporary uses of the em and en space?

In practice, em is the larger dash, en is the smaller—that is what they have come to mean and how each relates to other characters in terms of size.

It is nice if the en dash can do double duty as the minus, as Indra points out, but I don’t always remember to make it so.

Martin Silvertant

I've always used tabs as Nick said. I've never had a need for an em space so far. Are there instances where you couldn't go without?

Scott-Martin Kosofsky

Having learned typography in metal type (foundry type and Monotype), I've always retained its conventions. Anglo-American points and picas are my "native" measurements, and being an American, I work with inches, feet, miles, acres and such for everything else.

At its best, American publishing tends to be more strict with its rules than are its British counterparts. In the U.K., there's no style guide as comprehensive as the Chicago Manual of Style, or publications that have the kind of singular influence over editorial style as do The New Yorker and the New York Times. (They don't have a Constitution either, come to think of it.) Em- and en-dashes are intrinsic to any sort of scholarly or literary publishing here. The em dash is used to set off statements that might otherwise be parenthetical, but only when their purpose is to provide emphasis or contrast. The unspaced en-dash is used to indicate spans including dates, e.g., 1685–1750 or "see pp. 125–133." It is incorrect to use the hyphen or spaced hyphen in its place. The 3-em dash is still the standard to indicate repeated names in bibliographies. The minus sign and the plus traditionally occupied the tabular figure width.

In most American foundry type, em and en dashes were true to their names: the full square and the half square, with no sidebearings. In Monotype and Linotype, they were often a bit narrower and it was possible in Monotype to move them to a row that would give them some small sidebearings.

I often use em and en spaces in InDesign when I need a quick and regular way to indent some runover lines without resorting to tabs. What's nice about them is that their proportions (full-square, half-square) always look just right. I set them up as key combinations, as I do with all the spacing commands I use regularly.

I hate to mention it in this crowd, but I often wish my type designer colleagues had more hardcore experience as typesetters of a wide variety of demanding material. What I'm referring to aren't just matters of some ancient world of print, but are as much a part of literary and business communications as ever before.

Martin Silvertant

What I'm referring to aren't just matters of some ancient world of print, but are as much a part of literary and business communications as ever before.

I actually get frustrated often every day because people keep using hyphens everywhere. I keep a visual diary where I take pictures of everything typographically bad I notice on the streets or wherever I go. I took heaps of pictures of the misuse of the hyphen.

Having said that, I still prefer optical dashes. It may deviate from the historical conventions, but one of the issues I have is that the em and en dashes tend to be too long, where the en-dash to denote a range looks a bit odd to me and the em-dash is simply too obtrusive. In fact, many people seem to find the em-dash too obtrusive so instead they use an en-dash with single spaces on either side. The em-dash in the font on this forum is either too short and fat or a tiny bit too long and could have larger side bearings. I think the trick is to properly divide text without making the dash too obtrusive. I think it's hard to do if you follow the old convention strictly.

John Hudson said the width of the em-dash is supposed to equal the body height of the type. I just checked Lexicon and its em-dash is smaller than the x-height. Its en-dash is only a tiny bit smaller rather than half the size. Ohh, I just noticed when I type an em-dash in Lexicon it actually shows an extended en-dash. I just opened the Glyphs window to check if it's indeed the em-dash and found out it's decoded in place of the underscore. Anyway, I inserted the em-dash and measured its length and it actually exceeds the body height by quite a big amount. The en-dash is less than half its size.

I also checked Calluna and its em-dash is longer than the ascenders but smaller than the body height. The en-dash is roughly half the size but it's actually a tiny bit longer. I would have expected Lexicon to follow conventions—especially since Bram de Does translated all his knowledge of typesetting into its design—but it doesn't. Upon first seeing the em-dash of Calluna I thought it did follow conventions, and yet it doesn't either. I also checked Adobe Garamond Pro and its em-dash is a bit bigger than the length from descender to ascender while the en-dash is exactly half the size of the em-dash.

So I checked 3 random fonts now and only one more or less follows the standard. So is it really that important? Two of the fonts I just checked consciously deviate from the standard supposedly because they (the designers) think optically it looks right.

I also noticed in Calluna the dashes are the same weight while in Lexicon the em-dash is lighter. It's like I said, either make a long em-dash with less weight or a shorter one with the weight of the en-dash to not have the em-dash become too obtrusive. I don't know if there's a standard for the length and weight of the hyphen, but I think often the hyphen is too long. I tend to make it a bit shorter, and I make it a bit thicker than the en-dash.