The book, by Michael Coe, is even better. One of my favorite non-fiction books ever.
I think this is a little misleading. It’s not like there was a sustained effort for hundreds of years, and Landa’s scraps of the alphabet were lost for most of that. It’s not like it took centuries of chopping away at it, and it would take just as long if we started today from scratch.
Just ordered it from Amazon. I’ll get to it as soon as I finish the Da Vinci book (which is thoroughly enjoyable). Thanks for the tip!
I have the Complete National Geographic DVD set and a while back made PDFs from lots of articles in the issues (it was a multiple-step “hack”–they didn’t give a simple way to save material.) One of the articles is a nice 20 page feature about the Behistun Inscription in the December 1950 issue. My PDF is no longer at the site I uploaded it, but here is a site were you can read the text of the article without the illustrations (the second article down–Darius Carved History on Ageless Rock.) It is worth giving a read–recording the Behistun Inscription was much more like a scene from an Indiana Jones movie than leisurely reading the Rosetta Stone in a museum.
H. Beam Piper (no relation 
) argued in his short story, “Omnilingual”, that sufficiently advanced societies have a built-in Rosetta Stone: science, because it has universal principles. Those universal principles can be used as the beginning of deciphering an unknown language.
His story is set on Mars, where there had been an advanced civilisation before the Martian seas dried up, etc. The Terran archaeologists find plenty of inscriptions, but can’t decipher them.
Then they are exploring a building in a learning centre of some sort, and one of them recognises a chart on the wall. It’s the periodic table. It’s arranged differently, but because they know what it is, they suddenly have a small Rosetta Stone, that helps them with the basic scientific structure: that symbol is “H”, that symbol is “He”, that symbol is “1”, that symbol is “2”, and so on.
And once they’ve realised the value of looking for scientific charts and tables, they’re on their way to translating the language.
That’s the idea behind the Pioneer Plaque.
If we assume the rule of thumb of Zipf’s Law, then in a large corpus, let’s say, half of the words will be hapax legomena. Ideally one can figure most of them out, but if it is the name of some non-extant plant or something (we are talking about a million years from now?), one may simply be out of luck.
You realize that the Latin alphabet is derived from the Greek and is not completely different, right?
If you see the word kudos, you might not have to be genius to think of κύδος.
Anyway, speculating more on the direct subject of the OP, while ancient spoken languages may have been as complex as modern English, ancient written languages were all probably more simple than modern written English, especially the ones that used logograms, ideograms, or “shorthands” (such as not writing vowels.) I really doubt that any of the ancient written languages had anything close to the vocabulary in written modern English.
Actually, the abundance of English synonyms might work in its favor. If you know that “dog”, “canine”, and “hound” are all synonyms, then you only need to know one of those root languages to figure out all three words. And if you happen to know two of the root languages, then you can be very confident that you got it right.
Er… think you’re in error, there. In Ancient Egyptian, for example, there were about 1,000 different hieroglyphs and their writing combined logograms, syllabic, and alphabetic elements. Then there was also a “cursive” form used for daily writing and the much more formal forms for tombs and monuments. It was fiendishly complex and modern English, for all it’s aberrant spellings and rule exceptions, is actually quite simplified and streamlined in comparison.
Cuniform evolved from pictograms/logograms to syllabic and even some abjad/alphabetic elements. Again, fiendishly complex which might be why it was entirely supplanted by other writing systems.
Writing systems that don’t include vowels are properly referred to as “abjads” and they are not “shorthands”.
But, to return to the start of this post, it seems early writing systems were MORE complex than modern ones (Chinese being a possible exception - can’t say for sure on that one), and most modern writing systems, for all their flaws, are more streamlined and efficient.
It’s not the alphabet, or lack thereof, that determines the number of words, though. Modern English is known for having a fat dictionary (yet I wonder if the “typical” educated English speaker using 10000-50000 words truly averages more words than an Attic Greek counterpart?)
A far-out system of writing may be an initial obstacle in itself, of course, just as extremely complex grammar can.
The plethora of English comic books would reduce the simplicity to a high school project.
There’s something my Lit teacher called “vocabulary shift”. A modern farmer and his ancestor will have a similar amount and complexity of vocabulary, but where the modern one can describe the different parts and attachments of a combine and would have to take a guess whether that horse-shaped animal that’s smaller than a horse and has longer ears is a donkey or a mule, the ancestor would be able to identify equines without a thought and describe them and their possible ills in great detail.
The greatest difficulty we had reading old texts was usually due to those shifts; words about animals, farming life or armors which peppered any text older than the early 1800s were alien to us. A few years ago, a friend of mine was laughing because a game referred to a piece of armor as a quijote. I said “well, with that name it should be on the legs; is it?”. “Uh? What, you mean it’s an actual word?” “Yes, Don Alonso Quijano took his knightly name from a piece of armor!” And yes, the virtual quijote was on the legs. The word and its original meaning are still in the dictionary, but nowadays most Spanish speakers don’t know about it, only about the literary character; finding it in a novel in the 15th century would have been common, finding it in one in the 21st is highly rare.
Yes, OED2 has 291,500 entries. I’d hazard a guess that every distinct word in every written ancient document put together doesn’t reach that number.
It’s quite a good story and well worth reading.
There are people who make English texts in alternative scripts, for fun. So if it’s only the Latin alphabet that’s gone, there’ll be plenty ofconstructed scripts to work off of, much more than any Rosetta stone.
Note that one of those constructed scripts is the International Phonetic Alphabet, which is also used for other languages andfor English textbooks. Recovering English would be an absolute breeze with that. “Challenging” wouldn’t even be on the horizon.
I’m trying to imagine a circumstance where the Latin alphabet is lost, yet there are enough extant copies of texts in English that someone is trying to translate them.
Thing is, the Latin alphabet is so widespread and used in so many ways it would require something like the complete end of civilization for it to be lost. And not just the end of civilization, there would have to be no refuges or monasteries or institutions where the old texts are preserved.
We can read ancient Greek and Latin and Chinese and Sanskrit texts because those writing systems have been in continuous use for thousands of years. How exactly is the Latin alphabet supposed to vanish, when the fall of the Roman Empire itself couldn’t do that?
Anyway, let’s disregard the plausibility of the scenario. It’s thousands of years in the future, English is an extinct language, and the Latin alphabet has disappeared. There might be inscriptions and such left on monuments but nobody knows what they mean, after the great Transmogrification of 3729 AD and then the Rectification in 4588 AD.
Anyway, it’s 5417 AD, and it’s a new age of discovery. Rummaging through the radioactive zones, a giant cache of incredibly preserved paper books is discovered in a deep underground vault, written in the ancient script.
Is the question: is the fact that these books are written in English rather than French or Spanish or Turkish going to make the books harder to decipher? Why would it? English has some quirks, and our current orthography preserves some archaic pronunciations that are no longer actually spoken, but that won’t matter when scientists in 5417 are puzzling over the texts.
What might help is if there’s an existing language in 5417 that derived from the language in the texts. Coptic is the obvious analogy for Ancient Egyptian. If no known extant language is derived from the ancient language that makes it much harder. Obviously there are lots of modern languages derived from Latin, and Latin itself despite being a dead language has been continuously understood for thousands of years, so translating Latin is easy. If you don’t have any extant languages related to English then it’s going to be much harder. But that’s not because English is a particularly abstruse language.
And English is related to more languages than most, so there’s a greater chance of a relative surviving.
Right, so if any Romance or Germanic language survives, you’ll be able to pick out all sorts of cognate words, and to a lesser extent for other Indo-European languages.