Mark W. Post, University of Sydney
If you have studied almost any European language, you will have noticed words that felt oddly familiar. French mort (dead) recalls English murder. German Hund (dog) is a dead ringer for hound. Czech sestra resembles English sister. No prizes for guessing the meaning of Albanian kau (OK, well – it’s actually ox).
You might have wondered: could these words be in some way related?
Of course, words can look similar for various reasons. Unrelated languages borrow from one another: consider English igloo, from Inuktikut iglu (house), or wok from Cantonese 鑊 wòk (frying pan). And there are plenty of sheer coincidences: Thai ไฟ fai resembles its English translation fire for no particular reason at all.
But the preceding sets of words actually are related to one another. They are cognate, which means they share a common origin in descent from a single ancestral language.
This now-extinct tongue was probably spoken somewhere in Eurasia as many as 8,000 years ago. Long predating the advent of writing systems, its words – and its name, if it had one – were never written down. Lacking such direct knowledge, linguists have therefore developed methods for reconstructing aspects of its structure, and refer to it using the label Proto-Indo-European – or PIE.
But how do we know Proto-Indo-European must have existed?
Shared ancestry of language
Our modern-day awareness of the shared ancestry of Indo-European languages first took shape in the Renaissance and early colonial periods.
India-based European scholars such as Gaston Coeurdoux and William Jones were already familiar with the ties among European languages.
But they were astonished to find echoes of Latin, Greek and German in Sanskrit words such as mā́tṛ (mother), bhrā́tṛ (brother) and dúhitṛ (daughter).
Such words could not plausibly be borrowings, given these languages’ lack of historical contact. Sheer coincidence was obviously out of the question.
Even more striking was the systematic nature of the correspondences. Sanskrit bh- matched Germanic b- not only in bhrā́tṛ (brother) but also in bhar (bear). Meanwhile, Sanskrit p- aligned with Latin and Greek p-, but with Germanic f-.
There could be only one explanation for such regular correspondences. The languages must have descended from a single common ancestor, whose ancient breakup led to their distinct evolutionary pathways.
Philologists from the 19th century, such as Rasmus Rask, Franz Bopp and August Schleicher, later systematised these observations. They showed that, by comparing and reverse-engineering the changes each descendant language’s words had undergone, the words of the lost ancestral language could be reconstructed.
These insights not only laid the foundations of modern-day historical linguistics, but also went on to influence Darwin’s conception of biological evolution.
The Proto-Indo-European family tree
Forming a family
Like a biological genus, the Indo-European languages became understood as forming a family. At their root was the PIE ancestor, while the descendant languages branched out (like species) to form a tree.
The Indo-European family includes Indo-Aryan languages such as Sanskrit and Hindi; Iranian languages (including Persian and Kurdish); Hellenic (including Greek and Ancient Macedonian); Italic (including Latin, Spanish and Italian); Germanic (including English, Dutch and German); Balto-Slavic (including Russian and Lithuanian) and Celtic languages (including Welsh and Breton), as well as Armenian and Albanian.
Extinct branches – attested only through written records – include Anatolian (Hittite) and Tocharian. Languages like Phrygian, Dardanian and Thracian seem likely to have been Indo-European, but are not as well attested in historical records.
But not all Indian or European languages are Indo-European!
Non-Indo-European languages of India include Dravidian languages such as Tamil and Telugu, while European languages outside the family include Basque, Georgian, Maltese and Finnish. Compare with the above table the quite different Finnish words jalka (foot), isä (father) and viisi (five).
Learning about those who spoke PIE
Reconstructed PIE vocabulary has also yielded insights into the lives of its speakers: what their culture may have been like, where they might have lived, and what may have fuelled their language’s diversification and expansion.
Roots such as *rēg-, (tribal) king, and *pelə-, fortified high place, suggest a martial, hierarchical society.
Speakers knew grain agriculture (*agro, field; *grə-no, grain), animal domestication (*ghaido, goat; *gwou, cattle), vehicular transport (*wogh-no, wagon; *aks-lo, axle), metalworking (*arg-, shine or silver; *ajes, copper or bronze), trade (*wes-no buy; *k(a)mb-yo, exchange) and religion (*deiw-os, god; *meldh, pray).
From such evidence, scholars such as V. Gordon Childe linked Proto-Indo-European to the Late Neolithic/Early Bronze Age Kurgan culture of the Pontic-Caspian steppe (today’s Ukraine and southern Russia).
More recent studies have (somewhat controversially) employed phylogenetic methods derived from evolutionary biology to argue for a PIE origin in Anatolia (modern-day Turkey), suggesting that agriculture was most probably the engine of Indo-European expansion.
Alas, spoken languages do not fossilise well. The actual words, ideas and identities of Proto-Indo-European speakers vanished into the air many millennia ago.
Yet patterns in their descendant languages preserve enough structure to enable us to manage at least a shadowy glimpse of them. The theories and methods pioneered through this work will continue to fuel research into the reconstruction of human ethnolinguistic prehistories worldwide for many years to come.
Mark W. Post, Senior Lecturer in Linguistics, University of Sydney
This article is republished from The Conversation under a Creative Commons license. Read the original article.
