When a Roman emperor updated the alphabet – and basically nobody listened to him.
A common response to my profuse public pontifications about the history of the alphabet is ‘could we add new letters?’
With the phonemically faulty state of English spelling in mind, the question is one that I enjoy; it’s heartening to see people considering the alphabet not as a fossilised artefact, but rather as a living system, which any present-day generation can update to suit their speech. My answer to the enquiry is an enthusiastic yes, and I add that new members of the alphabet would draw strength from many historical precedents.
Over their millennia-long development, our letters have in various times and places given rise to orthographic offspring. They often emerge gradually. Different functions and shapes can conspire to create two letters where there was only one before. J and U are examples within the English rendition of the alphabet, having emerged from I and V since Roman times.
Inversely, two existing letters can coalesce into one new one. For an example of travel in this direction, see the letter W, the German ß, or even the Dutch IJ. The latter two illustrate how letter-hood is not a simple status, but a composite of usage, appearance and recognition. They, like W, spell a single sound and were born of two separate letters. Yet they still labour under the popular perception as a duo, and so envy W its secure position in the alphabetical order and its dedicated chapter in the dictionary.
Other letters, meanwhile, have been less organic in their origins. It remains an option for writers to create and start using a brand new character. Unlike the gradually emergent letter, we can expect such an obvious and sudden addition to the alphabet to face a steeper climb towards general acceptance. No matter how urgent the problem of sound that it serves to solve, that letter’s uptake will depend on factors beyond the linguistic – how it looks, how it feels to write, how entrenched the current alphabet is in writers’ minds, and who is arguing for the new letter’s adoption.
To test the final point there, we can pivot to one end of the social scale: how about having a Roman emperor for a patron? Surely such an august figure, an autocrat of antiquity, could guarantee on leaving a lasting mark on the alphabet?
Apparently not, says history.
If you’ve read or watched I, Claudius, you’ll know that the Roman emperor Claudius (10 BC–54 AD) was something of a swot and a scholar. Before being unwillingly elevated in 41 AD, burying himself in books may well have kept Claudius alive, while other members of the Julio-Claudian dynasty buried each other for real. Clearly a man of taste, his interests included the culture and language of the Etruscans; he is reported to have written an Etruscan history and dictionary, both now lost, much to the anguish of historians and historical linguists.
During his tenure as Rome’s princeps, Claudius not only provided the empire with sanitation, medicine, education, wine, public order, irrigation, roads, a fresh water system, public health and peace, but also found time to reform the alphabet. The writing system, approaching two thousand years old in Claudius’ day, comprised twenty-one letters, pruned back to spell well the sounds of spoken Latin. It ran from A to X, with the Greekish letters Y and Z hovering on the margins. Literacy was flourishing; the alphabet had already percolated through Roman society, from senators and scholars to the scatological scribblers of Pompeii.
Claudius thought, however, that there was room for improvement. His interest may have been personal as well as practical and intellectual; he belonged to the same clan as Appius Claudius Caecus, a reformer of the Republican era, whose successful campaigns are recorded to have included the removal of the letter Z from the alphabet. Latin at the time didn’t need it. Centuries later, when Greek acquisitions had made Z useful at last, the letter found its early position in the alphabet’s order had been awarded to G, so Z had to shuffle off to the very end.
Perhaps Claudius felt that alphabetical reform honoured his illustrious ancestor when he too fulfilled the function of state censor in the year 47 AD. Whatever the emperor’s motivations, later biographers, like Tacitus and Suetonius, include the detail that the emperor deliberately created new letters. Here is Tacitus on the subject:
“… ac novās litterārum fōrmās addidit vulgāvitque, compertō Graecam quoque litterātūram nōn simul coeptam absolūtamque … et fōrma litterīs Latīnīs quae veterrimīs Graecōrum. Sed nōbīs quoque paucae prīmum fuēre, deinde additae sunt. Quō exemplō Claudius trēs litterās adiēcit, quae ūsuī imperitante eō, post oblitterātae, aspiciuntur etiam nunc in aere pūblicō […] per fora ac templa fīxō.”
‘… and he added and made public new kinds of letters, having learned that the Greek creation of letters was also not completed in an instant … and the shape of Latin letters is like the oldest letters among the Greeks. But for us also, there were too few at first, so some were added. Upon this example, Claudius added three letters, which were in use while he ruled, but were latter forgotten about, although they can still be seen in public bronze tablets fixed in forums and temples.’
(Tacitus, Annals 11:13-4)
The emperor’s biographers agree that Claudius created and promoted new letters; that there were three of them; and that the three didn’t last long after his death in 54 AD. Tacitus makes a nice lexical choice when he describes their later state as oblitterātae – from the verb oblitterāre ‘to blot out, erase, obliterate’, itself possibly a fusion of oblinere ‘to smear over’ and littera ‘letter’.
The great thing is, we don’t have to take these ancient writers (with their many axes to grind) at their word. Archaeology confirms the Claudian update!
Okay, now I must qualify this article’s subtitle; the word basically might unfairly downplay the success of the three Claudian letters. Clearly, they did enjoy a degree of popular employment. We find them in stone public inscriptions and the remains of personal documents, in a social and geographical distribution that went beyond imperial eyes. That is to say, they turn up in some texts from Italy (but no further afield) that Claudius was unlikely to see.
Tremendous variation can be seen in the Latin writing from the Claudian era, with the new letters occasionally present in their appropriate function, but most times absent, and sometimes variably present and absent across the same text. This fluctuation, however, doesn’t erase Claudius’ achievement; some people were indeed doing what he asked. In a masterful PhD thesis on how Classical Latin was written down, Dr Veronika Nikitina speculates about one of the new letters’ vibes and the reasons for adoption:
“Perhaps the main factor for applying the new letter was the socio-linguistic factor of prestige associated with the new letters in Claudius’ time. This could be the main reason for the use … [in] the Forum of Augustus where the emperor himself might see it. In the inscriptions that feature freedmen and slaves the new letter could also be seen as a means to add prestige to their status by showing awareness of the new reform and to show their loyalty to the emperor. This could be especially relevant for slaves of the imperial household…”
(Nikitina 2015: 117)
Now then, what were these letters?
1. Ⅎ
Yes indeed, we begin with what is basically an upside-down F. To be accurate, this letter was inspired by the (now lost) Greek letter Ϝ, called digamma. Ϝ stood for the consonant /w/ (as in water), which Greek spelling used until the various accents of Greece dropped the /w/ sound. The letter was employed long enough to join the alphabet’s journey to Italy, where it would become our F. Academic knowledge of Greek digamma clearly survived longer still, as it inspired Claudius to pen the letter Ⅎ.
The Claudian digamma is the only one of the trio whose shape and sound we feel sure of. As with the Greek prototype, Ⅎ stood for the consonant /w/ or something close to it in first-century spoken Latin. For example, we have the verb vīxit ‘lived’, pronounced ‘weeksit’ and therefore spelled ℲIXIT on a Claudian-period monument. We also have DIℲI for dīvī ‘divine’ and PRIℲATIS for prīvātīs ‘private’. Personal names could be digamma-fied too, e.g. MINERℲA for the goddess Minerva and ℲITELLIUS for the consul and future emperor Vitellius. That said, its usage was far from thorough. A single document might write Ⅎ in one place, such as the title, but stick with the traditional V elsewhere.
2. Ⱶ
The shape of the second Claudian letter is known, thanks to a plentiful supply of archaeological finds, in which it may or may not be accompanied by Ⅎ. The sound that it stood for, however, is not known with any confidence. It was at least a vowel sound, one pronounced with the tongue high-ish in the mouth.
The not-unreasonable theory that you might encounter in textbooks and brief comments on the matter is that Ⱶ stood for a sonus medius, a ‘middle sound’ of Latin, with the high tongue held neither forwards nor backwards. It sounds a bit like English’s stereotypical schwa sound or the Russian Ы. Whether this vowel existed is one matter, but the question is made doubly complex by the observation that the Ⱶ is confined to Greek names and words taken by Latin from Greek. Given the lack of application to native words, we can infer that the sound behind the letter was also an import into Latin, a vowel approaching the status of a ‘loan phoneme’ (or ‘loaneme’, if you will).
Specifically, Ⱶ transliterates Υ (upsilon) in those Greek words and personal names; that letter served the high front vowel /y/ in Greek speech. Thus, we find the Claudian spelling AEGⱵPTI for Aegyptī ‘Egypt’. This restriction is strange, because Y was already on the scene and spelling that vowel for the Romans – as it still does in English: cycle, hyper, physical, type, etc. Why would Claudius not ally with Y?
The phonetic function of Ⱶ remains unclear for now. Nikitina suggests instead that it actually had nothing to do with a specific sound:
“I believe it is possible that the reform was purely orthographic: the letter could be originally introduced with the aim of eliminating spelling variation i/u/y, which occurs in Greek words and some remnants of i/u variation in Latin words…”
(Nikitina 2015: 116)
‘Citizens! Know ye not when to write I, U and Y in certain tricky words? Our emperor has, in his munificence, ingeniously solved your confusion…’ or words to that effect.
3. ?
Lastly, we have the doubly uncertain Claudian letter, in both its shape and its sound. To date, an archaeological example of it remains unfound. On the basis of a brief mention by the grammarian Priscian, it’s widely believed to have been Ↄ. Priscian, writing centuries after Claudius’ lifetime, doesn’t provide the shape, but does mention that its name was antisigma, the Wario to the Mario of the Greek letter sigma (C). On this basis, we can envision the mirror image. He also notes that it was intended to stand for two sounds: the pairs of consonants /bs/ and /ps/.
If correct in this, then Claudius wanted a Latin word like urbs ‘city’ to be written VRↃ. His own cautiously plain title, simply meaning ‘first man’, would become PRINCEↃ. Such a bi-consonantal letter wouldn’t have been so strange. The Greek had a perfectly serviceable equivalent in Ψ (psi), and our own quirky Roman letter X also conceals a sequence of sounds. None of this, mind you, is surer than speculation.
These are the Claudian letters: a trio of orthographic ghosts once pushed on the Roman public by their nerdy ruler. They enjoyed a few years of success while Claudius lived, then passed out of sight and mind. I like thinking about the generation of readers and writers subsequent to the Claudian period, those who occasionally encountered the old letters in old public texts. I imagine that the symbols stung the eye somewhat, as the idiosyncrasies of Victorian spelling may do for English-scanning eyes today.
So, why didn’t they take off? Why don’t the English alphabet and its likewise Latin-derived siblings now make space for Ⅎ, Ⱶ and Ↄ? In recent interviews, I’ve used the case of Claudian letters to make a point about the social circumstances of the alphabet – that its adaptation and modification over history is not just a matter of language, but is always caught up in the politics, culture, technology and education of a given era. I compare Claudius, a full-blown emperor, with Spurius Carvilius Ruga, the teacher and former slave who is credited with inventing the successful letter G centuries before. Yes, such stories are to be taken with a heavy dose of salt (we know that Appius Claudius Caecus didn’t also invent the letter R), but the Ruga legend is still plausible.
I compare the two to make the point that the alphabet could be easily modified in Ruga’s less-lettered day and age, but not in the later era of the Caesars. In the three intervening centuries, Latin literacy took off and the circle of writers widened, thereby making reforms and updates harder and harder to achieve. With each new generation, lasting contributions by individuals become exceptional, but the exceptions do sometimes happen. Sejong the Great, king of Korea and the project manager behind the Hangul script, comes to mind.
It’s a good point to make; I still stand by it. But having now spent more time with the Claudian letters and their context, I can’t help but tilt my thoughts towards an additional, maybe superior, explanation for why the trio didn’t outlive their inventor. It’s a linguistic explanation, separate from considerations of practice and prestige: perhaps they just weren’t good letters?
Their contribution to spelling and the problems solved by their inclusion aren’t immediately obvious. Yes, Ⅎ would provide the consonant /w/ with its own letter, but the alternative system of using V for that consonant and for the vowel /u/ worked just fine. The two sounds are in the phonetic relationship of a semivowel and its counterpart vowel; the same relationship operated behind the Romans’ use of I for the consonant /j/ and the vowel /i/. I recall no Latin words in which the pronunciation of a given V would be a matter of prolonged confusion. The addition of Ⅎ may have been a step towards a perfect ratio of sound-to-sign, but the people weren’t crying out for it.
The advantages of adopting Ⱶ and Ↄ were also minimal. The former may have ‘solved’ occasional confusion between I, U and Y by adding a further letter into the mix; the latter gave an occasional combination of consonants its own letter. Considered so, the failure of the Claudian letters seems unsurprising. Power and prestige are weighty factors in determining spelling, no doubt, but an equally powerful force for change is whether an update offers the benefits of linguistic clarity to the individual writer or reader. They each had to weigh those benefits against precedent and habit, and spell accordingly.
Like mushrooms served by Agrippina the Younger, this story is all starting to smell a bit funny – a tale of a bookish emperor unhelpfully meddling with Romans’ writing. Perhaps the citizens only went along with the pretence of his magnificent new letters when the emperor was passing by.
END.
Resources:
- Priscian’s Institutiones
- Tacitus’ Annals
Non-linked references:
- Bücheler, F. (1915). De Ti. Claudio Caesare grammatico: commentatio philologica. Kleine Schriften, 1, 1–4
- Nikitina, V. (2015). Standardisation and variation in Latin orthography and morphology (100 BC – AD 100). PhD thesis. University of Oxford.
- Oliver, R. P. (1949). The Claudian Letter Ⱶ. American Journal of Archaeology, 53(3), 249–257.
- Suárez-Martínez, P. M. (2024). Fonología de las letras de Claudio. MINERVA: Revista de Filología Clásica, 37. 29–43.
- Wallace, R. (2011). Latin Alphabet and Orthography. In Clackson, J. (Ed.) A Companion to the Latin Language. Blackwell Publishing Ltd. 9–28.
Images my own, from Wikimedia Commons or from credited sources. Cover image: bust of Claudius. Only human intelligence (or lack thereof) was used in the creation of this article.
Leave a Reply