What’s With the Diæreses

You may have noticed in my other writing, such as commit messages, that I use a diæresis on vowels, such as “coöperate” or “reärranging”. This is not a common pattern in written English today – the only word commonly spelled in this manner is “naïve”, and even it is often spelled with only one tittle1.

This habit is not the “metal umlaut” – the practice of putting umlauts over vowels to look cool or (European) exotic, like the metal band Mötley Crüe or the ice cream company Häagen-Dazs. It is an explicit recognition of the fact that the English spoken language has more than five vowel sounds, but only five vowel characters.

In English, when two vowels neighbor each other, they form a diphthong, and the spoken sound is understood to be neither of their individual sounds, but a different vowel, generally “in between” them. However, this runs into trouble when a different English practice mixes with the diphthongs: concatenation.

English uses the Germanic principle of modifying words by attaching a prefix or suffix stem to them. Anyone can act on their2 own initiative, but to act in response to external stimulus, they react. However, the word “react” does not rhyme with “reach”, even though they differ only in their final consonant (which is a digraph, not a diphthong – the “h” after a consonant forms a consonant cluster, representing a phoneme that is neither a diphthong nor a hiätus). The “re” prefix, while not a word, is an independent stem that fluent speakers know does not merge with the root word it modifies.

When spoken, the words “reäct” and “naïve” and “coöperate” do not flow from the first vowel smoothly into the second; rather, the speaker transitions abruptly, and the two vowels are in separate syllables. The diæresis (not an umlaut) marks explicitly that the word uses a hiätus rather than a diphthong. This may appear to be mere pedantry, but it does have important disambiguation properties!

Consider the following two sentences:

The worker coop (pronounced /kuːp/) was full of bustling activity.

The worker coop (pronounced /koʊˈɒp/) was full of bustling activity.

These sentences are context-sensitive, in that you must know details about which “worker coop” is being described here to know whether the word is coop as in cubicle farm or coop as in employee-owned it contains chickens or socialists (credit to Manish Goregaokar).

Placing a diæresis on the second “o” of the second “coop”, making it “coöp”, removes this context-sensitivity, and presents an unambiguöus3 sentence to the reader.

I, like the New Yorker, use the English diæresis to favor clarity and unambiguïty in language, and to reject the sociëtal trend that reduces the expressiveness of the alphabet and compresses the language into fewer and fewer orthographic symbols.

The advance of the emoji as a modern ideögraphic language is a case study in the human desire for disambiguäted glyphs and tolerance for a combinatorial explosion of orthographic space. If we felt the need to expand the emoticon “:p” into two different glyphs in order to distinguish between 😋 and 😛, then we can add a small handful of letter modifiers (and letters! Note that I’ve been using the ligature “æ” instead of the diphthong “ae” to spell the word “diæresis” for this whole article4) into our writing system.

My aims to reïntroduce þe θorn and θeta letters as disambiguätors for þe two pronuncations of the digraph “th” (and spell “digraph” as “digraf”) can wait.