Reconstructing Historical Accents

Would it be theoretically possible to reconstruct how people in various historical time periods spoke? For example, would George Washington have spoken like a proper British gentleman, or would a Virginian drawl come closer?

My hunch is that we couldn't begin to imagine how people spoke for the majority of human history. If the various American accents sound so different from each other, and American accents writ large from accents back in Britain, obviously a lot of linguistic evolution has transpired in the few centuries that have elapsed since the dialects' divergence. Odds are that Washington would have sounded like no one living today, British or Virginian.

That said, I do find it fun to imagine people from various historical cultures speaking with recognizable accents. For example, I like to imagine the Republican Romans talking like the Sopranos or the ancient Egyptians like Rafiki from The Lion King.

EDIT: Actually, this comes closer to how I imagine the ancient Egyptian accent.

edited 3rd Mar '13 7:30:54 PM by Jabrosky

Unless there's some kind of recording device or direct symbol->vocal chord movement translation somewhere, I don't think you could reconstruct it.

If there is a rich written corpus, you can examine which misspellings are more common. And if there is a poetic corpus and you know which rules the poems are supposed to follow, you can extract some information about the accent.

It involves a lot of hard work and guesswork, but you can generally find some solid information — for example, in the case of English, the great vowel shift (a.k.a., the main reason why English spelling does not make any damn sense).

English and other languages whose poetry includes rhyming can be rather accommodating that way. Also, the late nineteenth century is pretty easy to infer if you can get your hands on recordings of older speakers from the 1920s. See, your dialect tends to finalize around your 20s, so if you have a recordings of 90 year olds in 1920, you can guess how people spoke in 1850 when those folks were only 20 years old.

Yeah, it's just such misspellings and rhyming evidence that have let us reconstruct a reasonable picture of several English dialects.

Just to take the most glaring example: our modern English spelling system. Given the way we pronounce our words, who would have written them this way? Only someone who actually pronounced them more or less the way they're currently written. For the most part (broad exceptions noted), our spelling became largely set in stone with the introduction of printing in England—i.e., the mid-15th century. Deduction suggests that Englishmen of that time actually pronounced, for example, the word "write" as "w'rEEtuh," and "laughter" as "lowCCHterr" (with a throat-clearing sound in the middle).

Other examples:

• Misspellings indicate that in Shakespeare's day, the [ai] or "long i" in words like "sight." "I," and "bride" were pronounced more like [əi]—rather than an "ah" sound rising to an "ee," it kicked off with a "schwa" sound.
• Such evidence also suggests that 18th-century Englishmen pronounced syllable-final "R"s the same way modern-day Americans and Canadians do: RP pronunciations like "lettuh" and "singuh" didn't become features of British speech until the early-mid 1800's.
• And as The Rape of the Lock famously implies, in Alexander Pope's day "tea" rhymed with "say" and "obey."

Although there'll always be some speculation involved, there's a surprising amount that linguists can reconstruct. For example, there's a decent consensus on the nature of Proto-Indo-European, despite the fact that there are zero written sources for it.

Here's a video by the OU about Shakespeare's English. The accent is a bit unusual but still very familiar. It sounds a bit like a cross between West Country and Yorkshire. Definitely more comprehensible than Chaucer or the Pearl Poet, who were writing a mere 200 odd years before.

Edit (OT): Try reading a full text in Fraktur, which was used in some parts right up until the middle of the twentieth century. It ain't fun.

edited 4th Mar '13 1:41:23 PM by KorKhan

I alwayſ imagine they had a huge liſp ſo they would ſtumble over thiſ letter.

That Shakespearean English video is really cool. Thanks for sharing.

It's interesting that Chaucer's English has this sort of Scottish/Norse sound.

IIRC, tea was pronounced "tay" because that's how it is pronounced in French and they got the habit of tea-drinking from France. Trying to think if there are any other -ea words in English. If so, were they pronounced differently too?

edited 4th Mar '13 3:51:45 PM by Hodor

I guess you've got "sea" and "flea", both of which have an "ee" sound. We'd have to see how they were pronounced historically. On the other hand, "great" still retains the "ay" sound.

In the case of "sea", I can believe that it used to be closer to the German pronunciation of "see", which is pronounced similarly to "say".

edited 5th Mar '13 12:17:38 AM by KorKhan

To my ears, I hear Frisian or Dutch mixed in with English. When reading Chaucer, I always find it easy to switch on to phonetic reading... with my old South African accent on very strongly, switched to more Afrikaansy vowel-sounds and stresses (like, say... if I were trying to put on a Bloemfontein accent for jokey purposes, and taken beyond normal). It seems to work. <shrugs>

edited 5th Mar '13 7:17:06 AM by Euodiachloris

The Old English word for "sea" was "sæ"—the OE letter æ was pronounced like the vowel sound in modern "sat" and "sap." So it's highly plausible that since the Norman Conquest, the vowel would've raised halfway to an "eh" or "ay" sound as an intermediate stage, before raising all the way to an "ee" sound.

Old English sounds a lot like Norwegian to me.

Before English spelling got regularized, and before London speech became the agreed-on standard, there were several regional varieties of Middle/Early Modern English that give fascinating glimpses into how the language could have evolved if some region other than the Thames basin had become influential. In particular, the Northumbrian and Scottish literary dialects flourished for a few centuries, pretty much independent of southern English influence.

For one good example, check out the Scottish poet William Dunbar—his "Lament for the Makaris" is one of the language's overlooked gems.

Interesting that you bring that up just as I mention Northumbria and English-speaking Scotland—before the Conquest, those very regions were overrun by Norwegian settlers. After a generation or two, they were speaking English like everyone else, but the century of jostling between Norwegian & English heavily influenced the latter, and gave it a lot of very basic words and features that we still have today. In fact, such basic words as the pronouns "she," "they," "them," and "their" are all borrowings from Norwegian!

So someone trying to reconstruct certain OE accents could do worse than listening to a bit of Norwegian speech.

edited 5th Mar '13 9:08:04 AM by Jhimmibhob

The Yorkshire accent is apparently rather archaic.

Another interesting thing to note is that the modern "Strine" (rural/stereotypical) Accent in Australia is, linguistically, the closest to how the Cockney population of the 18th century sounded like.

This is due to how reletively isolated Australia is from the rest of the world, and the fact that mass migration only really happened in the late 19th century and then only really in the major cities sorta "froze" the dialect in time.

edited 10th Mar '13 2:57:36 PM by PippingFool

Thought I'd share this one, as well as this

Very interesting. From what I've read, grammar and syntax of proto-Indo-European (as opposed to the vocabulary) are still very conjectural. Still very cool listening to a text written in the language. For parts of it, you can even recognise the connection with modern descendant languages.