A post from Shady Characters

Miscellany № 34: what Ћ?

Those innovators who have designed new typographic symbols make up an eclectic bunch. Interrobang creator Martin K. Speckter was an ad man by trade and a printer by temperament; Bas Jacobs, designer of the ironieteken, and Choz Cunningham, creator of the snark, are type designers; and Doug and Paul Sak of Sarcasm Inc., responsible for the much-maligned SarcMark©, are an accountant and engineer respectively.

To this list — and I never thought I’d write this — we can now add ‘restaurateur’.

The big story this week is the proposal by Paul Mathis, an Australian restaurant owner, to use the symbol ‘Ћ’ as an abbreviation for the word ‘the’. As reported on the website of Australian newspaper The Age,

“The word ‘and’ is only the fifth-most used word in English and it has its own symbol – the ampersand,” says Mathis. “Isn’t it time we accorded the same respect to ‘the’?” […] He has developed the typography – effectively an upper-case “T” and a lower-case “h” bunched together so they share the upright stem – and an app that puts it in everyone’s hand by allowing users to download an entirely new electronic keyboard complete not just with his symbol — which he pronounces “th” — but also a row of keys containing the 10 or 15 (depending on the version) most frequently typed words in English.1

Mathis’ invention (although more on that slippery term later) has made quite a splash, reminiscent of the ripples caused by the SarcMark back in 2010, with many national newspapers and prominent websites picking up the story.2345 Like the Saks, Mathis has covered all his bases: the ‘Ћ’ has its own website (thethe.co), a Twitter account (@thefortweeting) and a promotional YouTube video. Most interesting of all, though, is the suite of Android keyboard apps — developed, Mathis says, at a personal cost of around AUD$38,0002 — with which adventurous users may type the ‘Ћ’ with the greatest of ease.

To my mind, however, the ‘Ћ’ is not without its problems.

Let’s start with a pet peeve. The ampersand (&), to which Mathis compares the ‘Ћ’,2 is derived from a complete word: it is a ligature, albeit a highly stylised one, of the word et. It literally means ‘and’, embodying the word in its entirety. ‘Ћ’ on the other hand, is a ligature of the letters T and h — the e in ‘the’ is left out in the cold. When, as The Age reveals, Mathis pronounces his symbol as “th”,1 he gives voice to this fundamental problem: as often as I scan the symbol ‘Ћ’, my brain persists in rendering it as a stunted “th” sound.

Secondly, it turns out that ‘Ћ’ is not a new invention. Though Mathis insists that he created the symbol from scratch (Karl Quinn of The Age gives him the benefit of the doubt1), he has had the misfortune of alighting upon a design that is functionally identical to the uppercase Cyrillic letter tshe, used only in Serbian, and which represents the <ch> sound in, for example, “chew”.6 Properly speaking, then, ‘Ћ’ does not even represent a “th” sound. (If there is an upside to this unfortunate coincidence, it is that many computing devices can be made to display a ‘Ћ’ without any special effort. Mathis’ Android keyboards are necessary only to enter the character, not to render it.)

Lastly, en route to his final design of ‘Ћ’, Mathis considered and then rejected an existing character — one that, unlike the Cyrillic letter tshe, has exactly the sound he was looking for. The ‘thorn’, or þ, is an Old English letter representing a “th” sound, and was once commonly paired with a superscript e as an abbreviation for ‘the’ to yield ‘þe’.7 Interviewing Mathis for American college website The Airspace, Blake J. Graham wrote:

[Mathis] looked back to the Old English thorn letter (þ) and its variant þe, which was used to represent “the” during the middle ages. While it seemed a good starting point, the thorn wasn’t the answer for Mathis. “Even in ye olde days these symbols were difficult to interpret and eventually were lost in translation,” he told me. “[It doesn’t] look like ‘The.’8

The irony here is that when Mathis talks about “ye olde days”, he is invoking the ghost of the thorn itself. ‘Ye’, commonly used as an anachronistic form of ‘the’ (“Ye Olde Tea Shoppe” is the canonical example), exists only because the cases of blackletter type that early English printers brought over from the continent lacked the Old English þ. Printers turned to the y, its nearest visual equivalent, and the thorn’s fate was sealed.9 Like the tshe, the thorn is well-supported by modern computers; unlike the tshe, the modern thorn boasts a well-realised form that is visually compatible with the roman alphabet. Might Mathis have dismissed it out of hand?

Despite all this, I have to admire Paul Mathis’ chutzpah in launching a new symbol, especially one that represents a word used so frequently in written English. Will the ‘Ћ’ stick? I can’t see it happening. Will his many Melbourne restaurants see a sudden surge in customer numbers as a result of worldwide coverage of his creation? I imagine so!

In other news this week, Symmetry Magazine tells the backstory behind Scott Fahlman’s creation of the smiley at Carnegie Mellon University back in 1982. I’ve talked about this before on Shady Characters, but read Julianne Wyrick’s article for some great details on the circumstances surrounding Fahlman’s invention.

For those who just can’t wait for the hyphen and dash chapters in the Shady Characters book, David Sudweeks’ article at FontShop about the usage of the hyphen, en dash and em dash will whet your appetite.

And lastly, a certain book helps Jimmy Stamp at the Smithsonian Magazine’s Design Decoded blog decipher “The Origin of the Pilcrow, aka the Strange Paragraph Symbol”.

Thanks for reading!





Subramanian, Courtney. “Why The Deserves a Symbol All Its Own”. Time NewsFeed.






Wikipedia. “Tshe”.


Scragg, D. “The Foundation”. In A History of English Spelling, 1-14. Manchester [Eng.]: Manchester University Press, 1974.




“Y”. In Concise Oxford Companion to the English Language. Oxford University Press, 2005.


33 comments on “Miscellany № 34: what Ћ?

  1. Comment posted by John Cowan on

    I don’t think it makes any difference that Ћ means different things in Cyrillic and (proposed) Latin: there are lots of letters with that property, like С (Cyrillic S) and for that matter Ѕ (Cyrillic DZ, now used only in Macedonian).

    1. Comment posted by Keith Houston on

      Hi John — and of course, many roman characters have different sounds in the various languages that use them. I’d still have preferred to see a genuinely new character, but at least it’s easy to display a ‘Ћ’here!

      Thanks for the comment.

  2. Comment posted by Andy on

    Maybe it’s just the font, but the lower part ‘Ћ’ looks too wide and the top part too narrow for my liking, especially compared to its component parts: T Ћ h

    I think the main problem with introducing new characters is simple: they’ll never take off because almost nobody can type them. A shame, but until a character appears as standard on a critical mass of keyboards (on-screen or physical), it’s doomed to obscurity :(

    1. Comment posted by Keith Houston on

      Hi Andy,

      Agreed! The problem is that ‘Ћ’ is a Cyrillic character that just happens to resemble a ligature of the roman letters T and h. It would need some typographic attention to bring it into alignment with the roman alphabet, but that would then make it look out of place in Cyrillic. A new Unicode code point would be needed to differentiate between the two.

      Paul Mathis’ Android keyboards are a valiant effort to kick-start the process of making ‘Ћ’ more widely available, but he’ll face an uphill struggle in displacing the standard Google keyboard and more popular replacements such as SwiftKey and Swype. (Full disclosure: I use SwiftKey, and it’s great!) Also, Mathis mentions in passing that Apple won’t even consider allowing a keyboard replacement app; it’s one of the inviolable parts of iOS. Thinking about it, perhaps the answer is not to shoehorn a new character into existing keyboard but rather to provide a keyboard with a Unicode search facility built in — users could type any character they want with relative ease.

      Thanks for the comment!

    2. Comment posted by Andy on

      Thinking about it some more, the rise of touch-based devices could actually improve the chances of new symbols being adopted, since obviously they’re not limited to having their layouts fixed in a physical medium.

      As you say though, it’s an uphill struggle to get people, even those folks who take the uncommon step of actually using something other than the default, to use a new keyboard. Getting the likes of Apple or Microsoft or the various third-part Android keyboard developers to include new symbols seems… unlikely at best, and so you end up with a chicken-and-egg situation: nobody uses the symbols because they can’t type them, and the keyboard makers won’t include the symbols because nobody uses them.

      Compounding the problem, I think symbols like these will only become commonplace if ‘normal’ people see a need for them. Considering how most people are content to use a hyphen in place of en and em dashes, and moreover given the general low standards for typed communications, I’m not getting my hopes up.

      And yet! Despite all that negativity, I still feel compelled to ‘do the right thing’ and use correct typography where possible, so I hope the… ‘inventors’? ‘developers’? of new symbols keep up the good work. It’s never been easy to predict what becomes popular and what fades into history :)

    3. Comment posted by Keith Houston on

      Hi Andy — I’m not so sure that it’s touch screen keyboards as such that will help new symbols gain a foothold, but rather predictive text. SwiftKey, for example, is almost preternaturally good at working out what I want to type next; combine this with a Unicode search facility and once I’ve entered ‘Ћ’ a few times it’ll start to suggest it without having to go through the whole search rigmarole. Of course, if this was combined with the ability to pin new characters to the keyboard, then so much the better!

  3. Comment posted by Andrew Areoff on

    How brilliant is this? I need to come up with my own symbol to represent at word.

    The only problem is, I don’t know if I can get these new characters in lead-type so I can set some letterpress cards with them.

    1. Comment posted by Keith Houston on

      I’m sure you could find someone to cast you a few ‘Ћ’ sorts. Hand & Eye in London periodically produce new fonts of type with their Monotype machine — perhaps they or someone they know could help you out?

  4. Comment posted by Korhomme on

    There already is a symbol for ‘the’. Pitman’s shorthand uses a ‘.’ .

    1. Comment posted by Keith Houston on

      Hi Korhomme — it’s funny you should mention that. As I was writing the post I wondered how one might go about abbreviating ‘the’ to make it as efficient as possible, and a simple point seemed most obvious. Thanks for the comment — I’ll have to read more about Pitman shorthand!

    2. Comment posted by Andy on

      How is it spaced? Does . symbol appear with spaces around it, as in . example I’m using here? I imagine. reader (not to mention auto-corrector) would get a little confused if. symbol appeared like a normal full stop.

    3. Comment posted by Korhomme on

      There is a space on either side. Pitman’s uses a small cross — an “x” — on the line as a full stop.

  5. Comment posted by Andy on

    The reference to typed communications in my earlier comment got me to thinking: perhaps a shorthand for ‘the’ will arise naturally from the way young people type – I can easily imagine how ‘d’ could become standard shorthand for ‘the’, just based on similarity in sound (bonus: it kinda looks like a backwards ‘þ’).

    1. Comment posted by Keith Houston on

      …or indeed an ‘eth’ (ð), which represents a softer “th” sound.

  6. Comment posted by Blimundus on

    Even if ‘the’ is used more frequently in English than ‘and’, it is much less important from a semantic point of view. Such a symbol would also be English-only. Some languages don’t even use articles.

    1. Comment posted by John Cowan on

      Not necessarily English-only. After all, & is not Latin-only despite its origin in et.

    2. Comment posted by Blimundus on

      Yes, but at least the Latin ‘et’ or the English ‘and’ can be translated in all languages using a single word, because it has a clear and single meaning. Contrast that with ‘the’ which basically means nothing more than ‘definite article singular’, which does not have a clear and single translation in a lot of languages (Dutch: de / het, German: der / die / das, French: le / la), and was not even used in Latin as far as I know.

      In Dutch, I am never going to use a symbol which, phonetically, sounds like ‘de’, for a word which would otherwise require a ‘het’ as article!

    3. Comment posted by Vikki McDonough on

      That would actually solve a problem faced by a lot of those languages (those that lack a gender-neutral animate third-person-singular pronoun).

  7. Comment posted by Damaris Wilson on

    You can call it chutzpah – I’d call it arrogance. My immediate thought was ‘what’s wrong with the ‘thorn’?’, quickly followed by – why not talk to some linguists?!

  8. Comment posted by Damaris Wilson on

    But I have to admit, it flows beautifully when writing it out – and saves space.
    I just felt that Mathis should have consulted some SMEs (go on, have a go …) before launching out :-)

    1. Comment posted by Keith Houston on

      I’m not sure I qualify as an SME! Having said that, I asked Erik Spiekermann (who is, surely, entirely deserving of SME status) on Twitter:

      Any thoughts on ‘Ћ’, Paul Mathis’ proposed abbreviation for ‘the’?

      To which he replied:

      I see no need. Ligatures are’t abbreviations, and special symbols like @ stand for more than a word.

  9. Comment posted by Graham on

    This interested me because, when handwriting, I’ve been using an e with a horizontal line above it for ‘the’ ever since I went to university in Zimbabwe for a term, where everyone seemed to do so as standard while not using any other formal or informal shorthand that I noticed. It’s quick to write and sort of makes sense as it looks like an e with the stroke of a t above it. I’ve been doing it for a decade or so and thoroughly recommend it! Not sure if it could/would catch on as a typographical symbol, although since the accent probably exists already in some languages perhaps it could?

    1. Comment posted by Keith Houston on

      Hi Graham — that’s intriguing. I don’t think I’ve heard of that before. Do you mean something like this: ‘ē’?

      Have any other Shady Characters readers come across this before, I wonder?

  10. Comment posted by themicksa on

    T’good folk of Yorkshire solved t’problem back in t’Dark Ages

  11. Comment posted by Wes on

    Ok, so the Cyrillic origins of the tshe, as well as the need for a new Unicode standard for Ћ on an Roman alphabet keyboard. Would concession also be made for the lower case form, ћ? For example: “Ћ maid was angry at me.” Versus “I made ћ maid angry”. Same sentence with the subjects switched around, but I wonder if the case difference between The Ћ ‘&’ the ћ would be necessary. After all, CAPS LOCK and Shift help all other letters. Or would Ћ just be another ligature like @ and & with no case variation? The only problem I see with that is that we rarely see “at” at the beginning of a sentence, and I distinctly recall a rule from first grade Language Arts that we never start a sentence with ‘and’ (‘And’). Yet we see ‘The’ there all the time. So to me, case variation is necessary.

    1. Comment posted by Keith Houston on

      Hi Wes — all good points.

      To me, given that ‘@’ and ‘&’ tend to be used in very specific circumstances, the lack of distinct upper- and lowercase variants isn’t a problem. For instance, it’s not just that sentences shouldn’t begin with ‘And’ (which is a fairly contentious rule at the best of times), it’s more that I wouldn’t expect to see an ampersand used in prose other than in deliberately idiosyncratic writing.

      Paul Mathis, on the other hand, wants to push ‘Ћ’ as a regular part of written language, and so, as you suggest, proper upper- and lowercase variants might be a nice touch. Unfortunately, to my eyes, the lowercase ‘ћ’ looks especially ugly — one more barrier to its wider adoption, I’d think.

      Thanks for the comment!

  12. Comment posted by Michael Hurley on

    A small correction for you. The thorn had started being written with a y shape well before printing came to Britain. As far back as the late 14th century you find examples of it. The below site has some examples scanned from original sources.


    1. Comment posted by Keith Houston on

      Hi Michael — thanks for the comment, and for the correction! I hadn’t known that the thorn had converged on a y-shape quite so early. Do you happen to know how fonts of type handled this? Had the ‘y’ replaced the ‘þ’, or was it merely that they were very similar?

  13. Comment posted by Zeissmann on

    ‘Þ’ has þe added benefit of being readily available from most standard US keyboards through þe AltGr+p combination. Just watch me type it nearly effortlessly: þ.

    1. Comment posted by Keith Houston on

      Hi Zeissmann — that’s odd; my laptop’s UK keyboard doesn’t seem to support that shortcut. Still, it’s good to hear that at least one country’s users can summon a thorn at will!

    2. Comment posted by Solo Owl on

      My I add a correction? The standard US keyboard does not have a key labeled AltGr. I know of no way to type Þ or þ on the standard US keyboard, short of keying in the Unicode number with the left Alt key, or using a Compose key if your operating system allows it.

      On the other hand, there is the US-International keyboard, nice bit of software that goes back to MS-DOS. Here you can type RightAlt+t to get the Icelandic thorn letter þ (RightAlt+p gives ö). To do this, press the Windows key, type “change keyboards” in the search box, and select the item at the top of the list.

      The Old English thorn letter, the one that looks more like a y than a p, is not available in Unicode. A few fonts have the glyph as an OpenType variant of Icelandic þ. How do I do that in a comment box?

    3. Comment posted by Keith Houston on

      Hi Solo Owl — you’re quite right; there may be no AltGr on a US keyboard. My understanding, however, is that RightAlt is functionally identical to AltGr, though as you say you may have to manually enable it to cause it to act as such.

      As for entering different variants of the thorn character, I’m afraid you’re out of luck! Webfonts allow websites to import whatever font they like, but comment boxes are typically (and deliberately) restricted to a small subset of the available HTML and CSS features that does not allow this.

      Your comment reminds me that I should really update shadycharacters.co.uk to use webfonts. At the moment it relies on a stack of common but not universal fonts to appear reasonably consistent across platforms; the lovely Hoefler Text, for instance, that Mac users see is not available by default on other platforms.

      Thanks for the comment!

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.