CIↃ in print

Andreas Stötzner · May 19

I try to chase specimen of this version of printing the number [Roman] M = 1000, which one finds sometimes on title pages. A quick search only got me two samples from coins (but the principle is the same).

image source: numista.com

michele casanova · May 19

Something like these?

Andreas Stötzner · May 20

yes thanks, this is what I meant.

Helmut Wollmersdorfer · June 21

It's the letter c turned 180°. BTW (I guess you know) the old Roman numerals have code points in Unicode.

https://en.wikipedia.org/wiki/Numerals_in_Unicode#Roman_numerals

What I also have seen in old printed books: IIII as number 4, or in blackletter typeface iiij with j as end form.

John Savard · June 21

Of course, though, IIII instead of IV for four was used fairly often in one other specific place: clock faces. Supposedly it's due to a superstition concerning the name of Jupiter.

Andreas Stötzner · June 28

Helmut Wollmersdorfer said:

It's the letter c turned 180°. BTW (I guess you know) the old Roman numerals have code points in Unicode.

In letterpress printing that C is rotated, of course, because you can’t mirror a lead letter. Therefore I wonder why U+2183(-84) is flipped, not rotated. Depending on the typeface, the visual difference may be hard to notice, but it seems odd to me nonetheless.

Image: https://us.v-cdn.net/5019405/uploads/editor/rl/13ke7fw35gzy.png

Andreas Stötzner · June 28

It seems we have a Unicode-typical mix-up of different things here:

Image: https://us.v-cdn.net/5019405/uploads/editor/xg/ditasypxsofd.png

U+2183 is defined as “roman numeral one hundred” and “Claudian antisigma” which appears as a clear error to me. The two are not the same. The proper solution would have been

to add the annotation “= roman numeral rotated one hundred” to U+0186
to name U+2183 “LATIN CAPITAL LETTER CLAUDIAN ANTISIGMA” and
group U+2183 together with U+2184 as a normal case pair; and
maybe put the Claudian letter pair not in the “Number forms” block but elsewhere.

b.tw., the term “Lowercase Claudian letter” is a bad joke, considered the historic presence of lowercase letters in emperor Claudius’ times.

I’d like to learn Peter Constables opinion on how this mess could get fixed.

John Hudson · June 28

Based on the samples seen so far in this discussion, it isn’t at all clear to me that this is normatively a rotated form as opposed to a flipped form. Yes, in some typographic examples, the C sort has been rotated, but that is not the case in the second of the coins you showed, Andreas, where the form is more obviously flipped.

Given that there isn’t a clear distinction in use between flipped and rotated forms, I think the Unicode unification of this form with the Claudian antisigma is probably fine. I would want to see a lot more non-typographic examples, before concluding that those are separate characters.

Qwerasd · June 28

By the way, this use of a backward C in roman numerals is generally known as "apostrophus":
https://en.wikipedia.org/wiki/Roman_numerals#Apostrophus

As for the design (flipped vs rotated): based on some late 16th–early 17th century examples (images and links to sources below), all of them look either ambiguous (due to unclear printing) or clear examples of the flipped (mirrored) C shape for the apostrophus, rather than a rotated C. It would be best, I suppose, to find type specimens where there is clearly a dedicated sort for the flipped C.

Do note that the glyph designs shown for characters on the Unicode code charts are not normative-- font designers are free to implement the design however they see fit.

If your aim is to imitate the appearance of the Ↄ from printed works where the printer did not have a distinct sort for it and was forced to rotate the C then you may freely design it as such- but if your goal is to imitate the appearance in many engravings and on many a plaque and in printings where a distinct sort for Ↄ was available, then you can opt for the mirrored option. You can also opt for the mirrored option just because you think it looks better, since it's well enough attested.

Images of Examples: