top of page
  • Paul Anthony Jones


(n.) the dot placed over the lowercase letters i and j

Why do we dot i's and j's? Lower case Roman numerals

Earlier on today, this fact cropped up on the HH Twitter feed:

It’s always nice to discover words for things you didn’t realise have names (we’re looking at you vartiwell, piqûre and manicule), and tittle undoubtedly falls into that category. But this great little fact raises a great little question: why do we even bother to put a dot over the letters i and j at all?

Actually, the second part of that question is much easier to answer than the first: J is just a modification of the considerably older letter I, and probably first emerged as a means of signalling the last figure in a row of Roman numerals—so 18 might once have been written “xviij”, 28 as “xxviij”, and so on. And because lowercase i had a dot, so did lowercase j. (Shameless plug: there’s more on that in our new book—which is now finally available in the States...)

Eventually, J was used enough in written language to warrant its own place in the English alphabet, and ultimately to earn its own distinct sound—more often than not, the voiced palato-alveolar affricate /dʒ/ (namely, the j sound found in words like jump, judge and janitor). But the fact that I and J were historically intertwined was enough for dictionaries as recently as the eighteenth century to continue lumping their I-words and J-words together. And it’s also the reason why Indiana Jones misspells “Jehovah” in The Last Crusade.

Crikey, what a film that was. Is it too late to make Marcus Brody the official mascot of Haggard Hawks? It is? Blast. But we digress.

So if j has a dot because i has a dot, how did i get its dot?

Well, when distinct lower-case letters first began to appear more than a millennium ago, they were typically just smaller, slightly simplified versions of their upper-case equivalents. This meant that lower-case i originally didn’t have a dot above it, because upper case I didn’t have one either. Instead, it was just a single, small vertical stroke or “minim”: ı.

This wouldn’t ordinarily have been a problem, except that at the time a great deal of written language was still being written in Latin, and because of that, an unexpected snag began to emerge.

Take a Latin word like foci, the plural of focus. Spell that with a dotless ı and it’s still perfectly legible, focı. But take a word like genii, the plural of genius—or for that matter radii, the plural of radius (or, according to Toyota at least, Prii, the plural of Prius). Change their two consecutive i’s to dotless ı and you’re left with genıı. Still legible? Well, maybe so in a nice, tidy, computer-generated typeface, but try to imagine it in handwritten cursive script, something along the lines of:


Suddenly those two ı’s become almost indistinguishable from a lower-case u:


(which has the unfortunate added consequence of being the Latin word for “knee”, not the Latin word for “intelligent people”.) Ultimately, Latin scholars faced a problem: the ii combination, which turns up fairly frequently in Latin vocabulary, could very easily be misinterpreted in these newly-emerging lower-case letters. Which brings us to this:

No one is entirely sure who made that sentence up (we’re giving generic “mediaeval scribes” the credit, but elsewhere it’s sometimes credited to a troupe of European actors), but the point remains the same regardless: a tangled forest of handwritten strokes or minims is, frankly, all but impenetrable. So to ensure that their texts remained legible, something clearly had to be done—and that’s where the tittle came in.

Starting not long after the Norman Conquest, writers, scholars and scribes began to add tiny dots or strokes above lower-case i (and hence lower-case j) to show without doubt that it was a separate character, independent from all those around it. The mimi numinum sentence above is thought to have emerged sometime later (probaby around the early thirteenth century) and would only have helped to popularize further this newly dotted-i by demonstrating, very neatly, just how imperfect the older undotted system could be.

Eventually this simple and ingenious solution became standard practice—and, like all the best solutions, has remained in place ever since.


Hi! We’re currently updating the HH blog, including all the tags (below). But with over 700 posts to reformat, well—apologies, this might take a while... 

For now, you can browse the back catalogue using all the tags from the blogposts we’ve already completed; this list will grow as more blogs are brought up to date.


Thanks for your patience in the meantime—and any problems or questions, just let us know at

bottom of page