Erected around 113 BC by an ambassador of one of the Indo-Greek kings and dedicated to the god Vasudeva-Krishna.
Allen also cites ancient sources (mainly grammarians) who discuss pronunciation, so this is probably also a source used in reconstructing ancient pronunciations of Chinese languages? Or is the work based mainly in analyzing predictable linguistic change over time and projecting back in time?
By the Sui/Tang Dynasty transition we have "rhyme books" (essentially phonological dictionaries) that detail every tone and phoneme for every character through successive dynasties, both through a variety of diagrams and categorizations, but also through a method called Fanqie (where every character's pronunciation is represented by two characters which separately represent the original character's initial sound and final sound).
While this yields all the phonemes of the language, we don't have their phonetic realizations from that alone (i.e. we know how many distinct sounds we have and their relationships to each other, but we don't know the precise, actual sounds quite yet). To find those, we use the usual other tools at the disposal of historical linguists which include:
+ Finding foreign loan words or transliterations and using the foreign language's pronunciation. A particularly fruitful source from this time period is Buddhist translations, which incorporate many words from Sanskrit.
+ Comparative reconstruction based on current Chinese varieties.
+ Analyzing what speakers themselves say about the phonetic realizations of the language.
By tracing this back to our first extant rhyme dictionary, called the Qieyun, with the attendant phonological system sometimes abbreviated as QYS for the Qieyun System, we arrive at what we call Middle Chinese (which is a concept that is fraught with danger as I detail later).
Although we still have some rather noticeable holes in our knowledge of the phonetic realizations of this system (the biggest is that we aren't certain of the exact phonetic realizations of three of the four tones of the QYS), we have on the whole a pretty solid understanding of the sounds of the language up to this period.
Going even earlier (i.e. Old Chinese) quickly becomes much more fraught with uncertainty without the aid of rhyme dictionaries. We are forced to rely much more on rhyming we observe in poetry as well as theoretical internal reconstruction based on our much more stable reconstruction of the QYS. By looking extensively at ancient poetry collections such as the Shijing and Chuci we can back out some system of phonological categories. As such these earlier reconstructions are far more contentious and uncertain.
The messier answer:
Middle Chinese (or Middle Sinitic as Mair calls it) was taken seriously as a language that people actually spoke and as the ancestor of Chinese varieties by the early Sinologist Bernhard Karlgren. This has led to terms such as "Early Middle Chinese" and "Late Middle Chinese" and trying to date those etc. This has changed significantly in the last 40-50 years. Modern Sinologists who study the evolution of the language basically refute every clause of what I just said. In particular,
+ Middle Chinese does not represent a language anyone actually spoke. It's an abstract system of phonological categories that represents a compromise among different varieties spoken at the time, resulting in more phonemes than any real human actually used in normal life.
+ As such terms such as "Early Middle Chinese" or "Late Middle Chinese" are linguistically incoherent, since there was never an organic language to evolve in the first place.
+ The traditional genetic, tree model of language families (with e.g. different descendant languages evolving from a single ancestor) is not a useful model for the evolution of Chinese, especially in the last two millennia. As such the whole enterprise of trying to find a most recent common ancestor for varieties of Chinese (or even subsets such as all varieties minus Min or all varieties minus Min and Wu) is methodologically suspect.
+ To the extent that you still try to find the most recent common ancestor for even non-Min/non-Wu languages, you are likely to arrive at a time period that far predates the QYS, and hence the word "Middle" no longer seems like an accurate chronological descriptor.
A new crop of terms has shown up to try to displace Middle Chinese and focus on trying to reconstruct a real language people actually spoke (e.g. "Medieval Chinese"), but even that is still quite messy.
The story of Old Chinese is even more contentious and is completely fractured in the academic community. Some Sinologists have staked their entire careers on a coherent notion of "Old Chinese" and its reconstruction. Other Sinologists dismiss the entire enterprise as unworkable, apart from potentially much more isolated and restricted subsystems. It's all very controversial.
The article touches on this, but when Chinese explorers arrived in Bactria (roughly modern Afghanistan), they were expecting to run into "ioanians", due to earlier texts that had described these people (大宛 - Da Yuan, meaning "great ioanians").
This is the early history of the silk road, and also the first appearance of Buddhism in the West, which must have been absolutely fascinating to Stoics.
Not picking nits, just an interesting fact. At that time, Asia was the name of the province on the eastern shore of the Aegean, where the Ionian Greeks were also located, so in a sense the sentence is absolutely true.
However the name "Ionian" still survives to this day throughout (modern-day) Asia: from the old Greek "Mikra Asia" in Anatolia (modern-day Turkey) all the way to India, Greeks are "Ionians": "Yunani".
Ah, and I just remembered another curious little thing:
The western island complex of Greece (you know, Corfu-Zante-woohoo! etc., plus Odysseus' Ithaca, whichever among the candidates it may be) are also called "Ionian", but traditionally were inhabited by Doric, rather than Ionian, Greeks.
I seem to remember anecdotes of the greek expedition bringing Cynics along, who seemed to really see eye to eye with the hindu "gymnosophists".
But I'd expect that two interacting merchants would definitely know of each other's names, as semi-contractual epistles of orders and corresponding shipping invoices were very much a thing (actually already since Sumer).
Not to mention that when you met someone back then, from what I understand, the idea would be that you would update each other on the stuff you knew about the world at large, sort of like a Fidonet where you got to walk to the next node: the time you spend with each other is precious, and ends up constructing a global grid of relationships and information.
It's mindblowing to consider these past ages, and how all those people who just went on with the business of living and learning even without GPS or Wikipedia, and did pretty well regardless.
They definitely had GPS back then, which has just barely survived into the 21st century: https://www.mongolianexperiment.com/mongolia/navigating-by-m...