A post from Shady Characters

Miscellany № 60: the secret life of the tilde

I am guilty of having somewhat ignored the tilde, or ‘~’, here at Shady Characters. Having just read Joseph Bernstein’s excellent, recent BuzzFeed article on the subject of “The Hidden Language of the ~Tilde~”,1 however, I thought I’d take a fresh look at this quirky sun-dried hyphen.

Joseph opens his article with a lament:

Last month in New York, Adam Sternbergh began his long cultural history of emojis2 by contrasting Face With Tears of Joy, the world’s most popular emoji, with the tilde, the venerable squiggle that is surfed on QWERTY keyboards by the ESC key and in math means approximately. Sternbergh pointed to the fact that Face With Tears of Joy has grown more popular on Twitter than the tilde as sufficient reason to offer tongue-in-cheek, if not Hearts in Eyes, advice to the ancient symbol:

“The 3,000-year-old tilde might want to consider rebranding itself as Invisible Man With Twirled Mustache.”

Unsettled by Adam Sternbergh’s abrupt dismissal of the poor old tilde, Bernstein proceeds to make the case for the tilde’s continuing relevance in the digital world, explaining that the ~bracketing tilde~ “unquestionably does something to [words], something destabilizing and a little uncanny […] a good definition of the use of bracketing tildes might go no further than adds juju.” Rather than recapitulate Bernstein’s arguments here, I thought a little context might be in order. Because when Sternbergh offhandedly mentions that the tilde is 3,000 years old, he is not far wrong — it may not be quite so ancient as that, but this little twiddle is the scion of a very long-lived family of marks.

Like the versatile octothorpe (#) and @-symbol, the modern tilde has a variety of uses. In mathematics it can be placed before a number to mean “approximately” (“~50” means “about 50”), or inserted between two variables to mean “x is equivalent to y”; in Boolean logic it stands for negation, or “not”; in tweets and other online discourse, as Joseph Bernstein says, it adds a sort of sarcastic or ironic emphasis to words;* and in the ubiquitous Unix-based operating systems on which the Web runs it is an alias for the home directory of the current user.

It is safe to say that its earliest adopters had none of these in mind.

One of the best clues to the origin of the tilde is to be found in the King James Bible, the “Authorised Version” that has lorded it over lesser translations of the Bible since the early part of the seventeenth century. In Matthew 5:18, as the King James version has it, Jesus says to his disciples:

For verily I say unto you, Till heaven and earth pass, one jot or one tittle shall in no wise pass from the law, till all be fulfilled.3

The KJB was translated from a patchwork of Greek, Latin, Hebrew and Aramaic sources, with this particular line taken from the original Greek text of the New Testament. In the Greek, “jot and tittle” are rendered as iota and keraia respectively, where iota is the name of ‘ι’, the smallest letter of the Greek alphabet, and where keraia means a small hook, serif or diacritic mark used to modify written letters.3 (The linguistically diverse sources for the KJB, however, do leave some wiggle room for interpretation. The small Hebrew letter yodh (י) provides a tantalising alternative to iota, and ancient Hebrew scribes were well aware of how a keraia could transform one letter into another — ‘ב’, or B, is distinguished from ‘כ’, or K, only by the subtlest stroke of the pen.4)

But if iota (or yodh) is straightforwardly mapped onto “jot”, then what is a “tittle” and why was it chosen as the English counterpart for the Greek keraia? The Oxford English Dictionary explains that “tittle” comes from the Latin titulus, for “superscript” or “title”, and that it refers to a whole host of accents or other marks placed above letters or words to change their sound or meaning. A medieval a modified by an accent to form á, for example, became the long form of the vowel; an a drawn with a line above it (ā), on the other hand, represented an abbreviation beginning with a. And though we might more correctly call the acute accent (´) and macron (¯) by their formal names, they are both very much tittles — as are the dots at the top of is and js. As obscure as it is today, “tittle” is as accurate a translation for keraia as anything else; by talking of jots and tittles, the translators of the King James Bible let Jesus allude to the finer points of law by analogy with the smallest marks made by a scribe’s pen.5

You may have guessed where this is leading.

In the Latinate languages that preceded modern Spanish and Portuguese, a dash or ‘~’ placed above a vowel indicated the omission of a following n or m — a so-called nasal consonant — so that, for example, aurum, or gold, could be abbreviated to aurũ.6 As medieval Latin evolved into the modern languages of the Iberian peninsula, these missing nasal consonants gave rise to the “mouillé” sound found in Spanish (in the word señor, for example) and the nasally-inflected ã and õ vowels found in Portuguese.7 Thus the tilde was and is a tittle par excellence, a mark used to modify the sound or meaning of a letter. It was so exemplary of the form, in fact, that the word “tilde” itself arose from “tittle” sometime during the nineteenth century.8 In word and deed, the tilde is a tittle, with roots that twine through classical Greek writing, Latin vocabulary and Iberian speech.

Bernstein, Joseph. “The Hidden Language Of The ~Tilde~”. BuzzFeed.


Sternbergh, Adam. “Smile, You’re Speaking Emoji”. New York.




Jones, Dr. Floyd Nolen. Chronology of the Old Testament. New Leaf Publishing Group, 2005.


OED Online. “Tittle”.


Sampson, Rodney. Nasal Vowel Evolution in Romance. Oxford University Press, 1999.


OED Online. “Tilde”.


Google Ngram Viewer. “Tittle,tilde”.


The tilde was even pressed into service as an explicit irony mark, as discussed some years back here at Shady Characters

11 comments on “Miscellany № 60: the secret life of the tilde

  1. Comment posted by John Cowan on

    In the online role-playing game I used to participate in, some of the characters had the power of perceiving the emotions of others. When one of them was on stage, other characters would use double-tilde brackets to delimit an unspoken but detectable emotion. ~~ desire to explain ~~ nostalgia ~~

    1. Comment posted by Keith Houston on

      That’s a nice way to handle that situation. For some reason, I’m put in mind of early Japanese CRPGs — if I remember rightly, Final Fantasy and its ilk used similar means to convey emotions or other aspects of the game that were otherwise too subtle for an 8-bit console to do justice.

    1. Comment posted by Keith Houston on

      Interesting! Neither of the sources I looked at talked about the iota subscript in the context of this passage, but a cursory Google search does bring it up. In the course of looking into it, though, I read that the Greek concepts of iota and keraia as suggestive of small scribal marks would likely have been unfamiliar to Jesus and his disciples, who would have been more familiar with Aramaic and Hebrew. The KJB may have taken some liberties with the translation, but the original Greek text was already some way away from being a direct transcript.

      Thanks for the comment!

  2. Comment posted by Korhomme on

    I remember that the Latin textbooks I used decades ago had a line over some vowels; this is a macron, I presume, and was used to indicate a lengthening of the vowel sound.

    Diacritics aren’t really used in English, outside poetry, such as ‘agèd’ rather than ‘aged’. And that’s one of the major problems in English and its spelling. We are accustomed to the difference between, say, ‘mat’ and ‘mate’. It’s the final ‘e’ that indicates the lengthened vowel sound. Try it with many short words ending in ‘e’, remove the ‘e’ and hear what happens.

    The basic problem in (written) English is that we have 26 letters in the alphabet, but we can produce between 42 and 44 phonemes, vocalisations for which there are no direct alphabetic correlations.

    This is one of the reasons that English spelling can be so difficult. Dr David Crystal has a wonderful little book on how spelling developed over the ages, well worth reading.

  3. Comment posted by Jamsheed on

    This is a lovely reminder of why I love this blog. Thanks!

  4. Comment posted by Dick Margulis on

    Just a note that the swung dash is not the same as the tilde, even though they’re reduced to one key on the keyboard (similarly to the hyphen-dash). The tilde is the diacritic over the the n in cañon. The math operator is the swung dash, ~, and despite the popularity of calling it a tilde, it’s a different glyph.

    1. Comment posted by Keith Houston on

      Hi Dick,

      The OED says this about the swung dash:

      a curved dash ∼, used in dictionaries to stand for the headword of an entry or for a specified part of it. In Oxford dictionaries first used in the first edition of The Little Oxford Dictionary (1930) but there called a tilde.

      Google Ngram Viewer suggests that the phrase “swung dash” surfaced around 1960 or thereabouts — I wonder if the swung dash is the tilde, only in its atomic form?

      Thanks for the note!

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.