A post from Shady Characters

Irony & Sarcasm marks: intermission

Adam Rice wrote in response to the discussion of ‘ironics’, or backwards-slanting italics, in Irony & Sarcasm marks, part 2:

I’ve occasionally seen reverse-italics on headstones dating from the 1800s, but I’m confident they’re not being used in a tongue-in-cheek manner.1

Adam was good enough to provide some examples; one can be seen below, and more are available in this photo set and others at his Flickr account. I have to agree with him — this is very definitely not an ironic use of ironics.

Reverse italics on gravestone, by Adam Rice
Reverse italics on gravestone. (Image courtesy of Adam Rice.)

Also, Kirill Grouchnikov mentions ironics in a post on Google+:

Keith Houston is back with another installment that provides an eclectic mix of history, gossip and typography, with a slightly jarring insistence on using the word “alphabet” for what is just a typeface.2

I’ve commented previously on the distinction between a simple variation in font style and an entirely new ‘alphabet’, but I’m no typographer. Can any Shady Characters readers shed some light on what makes roman and italic letters different alphabets, and whether ironics qualify or not?

Anyway, thanks again for all the comments, tweets and posts, and check back on October 2nd for Irony & Sarcasm marks, part 3!

Rice, Adam. “Comment on Irony & Sarcasm Marks, Part 2”. Keith Houston, September 2011.


Grouchnikov, Kirill. “Post on Google+”. Kirill Grouchnikov, September 2011.


16 comments on “Irony & Sarcasm marks: intermission

  1. Comment posted by Jason Black on

    I’d say, based on dim recollections from college courses, that an alphabet is a conceptual, abstract entity, while a typeface is concrete entity.

    An alphabet is a set of abstract glyphs which, collectively, represent the phonemes of a language in a manner sufficient to represent that language in visual form. One can probably be more pedantically rigorous than this definition, if one wants, but the important bit is that an alphabet maps most closely to the language’s phoneme set than to anything else. (Contrast “syllabary, for example.)

    A typeface, then, is simply one manifestation of those abstract glyphs in concrete terms of the exact shape of each glyph.

    Or in other words, the alphabet is a set of answers to the questions “what makes an ‘a’ an ‘a’?” and so on for each glyph in that alphabet. A typeface is simply a collection of specific glyphs, each one of which fits into the visual space covered by those answers.

    Or something like that, anyway…

    1. Comment posted by Keith Houston on

      Hi Jason,

      I think you’re right — it seems reasonable to say that an alphabet is a set of abstract glyphs and that a typeface is a concrete realisation of those glyphs. I could have sworn I read the term ‘italic alphabet’ somewhere recently, and the concept lodged in my mind such that I applied it to ironics as well.

      Thanks for the detailed response!

  2. Comment posted by Geof Huth on


    A typeface is actually a collection of fonts, a single family but with different views of each. So the italics are a font in a typeface, just as is bold and small caps, and sometimes many others. Hope this helps. But I don’t much care. I found the use of “alphabet” a bit jarring, but who didn’t know what you meant?


    1. Comment posted by Keith Houston on

      Hi Geof — I do occasionally commit the cardinal sin of conflating ‘font’ with ‘typeface’, so I must apologise for that. Thanks for the comment!

  3. Comment posted by Mark on

    Regarding back-slanted type, there’s also the question of which way to slant letters in a right-to-left font. Hebrew isn’t entirely suited for italics—but that doesn’t stop people from making them anyway. I have seen Hebrew “italics” slanted both rightward and leftward (rightward more often). See http://new.myfonts.com/fonts/adobe/adobe-hebrew/italic/glyphs.html for example. Hugh J. Schonfield’s bizarre reformed Hebrew font (visible at http://web.archive.org/web/20091025055638/http://geocities.com/snortar/schonfield.html) included a leftward-slanting italic (note the rightward-slanting “normal” (Frank-Ruehl) Hebrew shown as an example above it).

    In general, I’ve found the Hebrew typography, in adopting and responding to modern typographic conventions, tends _not_ to mirror-reverse asymmetrical glyphs and such. Obviously, such things as parentheses are exceptions (Sorta. In Unicode, “(” is always the open-paren, even in Hebrew, and it’s just drawn the other way around. Or you can consider it to be using the parens in the opposite order). But question-marks and commas, from what I’ve seen in print, are generally left alone, which means that the question-mark opens _ahead_, not looking back on the sentence it’s closing, and the comma similarly does not look like a small “closing” mark.

    (Arabic’s style is so cursive it’s sort of pointless to talk about “italic” for it, though I’m sure there are obliques out there. Arabic’s question-mark (؟) _is_ reversed to look back on its sentence, and the comma (،) (as well as the semicolon (؛)) is large and actually inverted, not just reversed.)

    1. Comment posted by Keith Houston on

      Hi Mark,

      It’s interesting that you should say that “Arabic’s style is so cursive it’s sort of pointless to talk about ‘italic’ for it” — given that italic and roman letterforms are themselves derived from handwritten letters, perhaps it’s just that we haven’t found an appropriate Arabic hand from which to derive alternative, ‘italic’ letterforms.

      Thanks for the detailed comment!

  4. Comment posted by Jon Ericson on

    I’m no typographer myself, but years of work with the LaTeX system of typesetting has trained me to call the difference between italic and roman lettering a difference in shape. Often these two shapes are paired to form a font family, though italics often pair well with a variety of roman (sometimes called normal) typefaces.

    Interestingly, font families sometimes include a slanted roman (or oblique) shape that serves as a sort of poor-man’s italic. According to Donald Knuth in “The TeXbook”, slanted roman type was introduced in the 1930’s, but the photo above clearly shows the shape’s use even earlier. (Line 3 is all-caps roman slanted, though it is carved not printed.) He also suggests that slanted and italics can coexist in mathematical (and by extension other technical) documents because italics are often used to represent variables in formulas.

    Another shape, commonly used for titling and often the first few words in a chapter, is small caps which uses smaller-sized versions of capital letters in place of lowercase letters. Theoretically, upright italics would be a further shape, though I’ve never seen them used in earnest. So if I were to implement ironics, I’d make them another type shape.

    The tombstone is remarkable since I see at least 4 different faces used: italic, sans-serif all-caps, slanted serif all-caps, and ironic serif all-caps. Modern sensibility would be to limit typefaces to at most two for display items such as titles, posters, signs and tombstones. The purpose of the rule is to minimize confusion about the intended semantics of the font shape. In this case, it’s not clear why the given names are in a different font than the surname or why “BORN” and “DIED” are reverse slanted.

    1. Comment posted by Keith Houston on

      Hi Jon,

      ‘Font shape’ would seem to be a sensible terms to distinguish between italic, slanted, small caps et al while retaining the meaning of ‘typeface’ and ‘font’. I used LaTeX at university (though that was a good while ago now) and I’ve always been intrigued by Knuth’s approach to digital typesetting. The concept of ‘badness’ alone would be well worth a blog entry.

      Thanks for the comment!

    2. Comment posted by Solo Owl on

      What is, or should be, the difference between italic and slanted capitals? The capitals in Donald Knuth’s slanted, italic, and math italic look the same to me, except maybe for very subtle differences in the bowls of his Js.

    3. Comment posted by Keith Houston on

      Hi Solo Owl,

      That’s a good point! I suppose one might use the term “italic capital” to refer to a capital that is slanted at an angle to match lowercase italic letters, as opposed to one whose angle matches that of slanted lowercase text — assuming there’s a difference in that angle, of course. And if we do distinguish between the two, we’re better placed to discuss them if there are any physical differences between slanted and italic capitals.

      (Incidentally, I’ve fixed the italic s at the end of the last comment. Don’t hesitate to get in touch if you’d like to have a comment edited after the fact!)

  5. Comment posted by Nate on

    I believe (and this is picked up solely through amateur font work) that the technical distinction is that Roman and Italic characters are ‘scripts’ — that is to say, different writing systems that cover the same letters. So alphabets are abstract collections of letters, scripts (and English has and has had other scripts, such as cursive, blackletter, and uncial) are a form that the alphabet takes (collectively), a typeface is a particular implementation, which may contain one or more individual fonts. But because a typeface is a created work, not a conceptual category, someone might create a typeface that includes fonts of different scripts — although usually they just include one script, with variants in weight and slant.

    Of course, there you get into the cranky distinction between “obliques” and “italics” — in theory, Italic was a separate script, but in common usage in the computer age it just means “a slanted Roman.”


    1. Comment posted by Leonardo Boiko on

      I come from a linguistics/writing systems background and (at least in my corner of the world) we call them “scripts” as well. Not many people realize that today we mix three scripts with separate histories: uppercase (from Roman capitals), lowercase (from Carolingian minuscule), and italics (from Italian chancery). All of those were originally different calligraphic scripts, and were used to write entire texts; but from the start copysts and typographers enjoyed mixing scripts in the same work for æsthetic effect, and in time it resulted in the present system (where each script has a well-defined role).

    2. Comment posted by Keith Houston on

      Hi Leonardo — thanks for the comment. I wrote a little about these various scripts — Roman capitals, Carolingian minuscule, italics and also blackletter — back in The Ampersand, part 2, though clearly I lost the finer distinctions of the terminology pretty rapidly after that!

  6. Comment posted by Richard Jenkins on

    It seems appropriate that the suggestions for implementing “ironic” type date from the early 20th Century as type families with matching roman, italic, and bold designs became common.

    I imagine this innovation grew out of the adoption of hot metal typesetting systems like Linotype. Rather than investing over the course of years in individual cases of loose type, the printer could buy a set of related typeface matrices for their system at the fraction of the cost. Presumably, font changes could also be accomplished more quickly by changing a magazine of type matrices than pulling out a case of loose type (and then returning it later).

    Once writers and editors became accustomed to getting italic and bold type set with no more effort than marking a simple underline or squiggle on their copy sheets, they could then go on to imagine new variations of the type family.

  7. Comment posted by Solo Owl on

    In the back matter of his book Computer Modern Typefaces, Donald Knuth provides examples of all his fonts. You can see Computer Modern Roman, Computer Modern Slanted, Computer Modern Italic, Computer Modern Upright Italic, …, some of them bold extended. As explained in The TeXbook and in Shady Characters, the difference between italic and slanted lies in the shapes of such letters as a, e, i, w, &c.

    There is also Computer Modern Funny Font, a backsloping italic face. It was an experiment designed to stress his software and not to be used. If you have Metafont, I suppose you can create your own Computer Modern Ironic just by changing a single parameter.

Leave a comment

Required fields are marked *. Your email address will not be published. If you prefer to contact me privately, please see the Contact page.

Leave a blank line for a new paragraph. You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>. Learn how your com­ment data is pro­cessed.