7 NORTHERN IDYLL Baltic and Slavic
In 1962, almost twenty years after fleeing her native Lithuania, Marija Gimbutas published a book called The Balts. It was a distillation of her encyclopaedic knowledge of the Baltic region, from its folklore and religions to its languages and archaeology, but she couldn’t quite keep her nostalgia out of it. She was, after all, an exile, having been forced to stay away from Lithuania by its post-war absorption into the Soviet Union. Sitting atop a Californian hill, with broad vistas in all directions, she saw in her mind’s eye the hills blanketed with oaks that surrounded Vilnius. ‘The Californian sand dunes, at Carmel, remind me of the pure white sands of Palanga, where I used to collect handfuls of amber; and the sunsets in the Pacific, of the peacefully sinking sun as it disappeared into the Baltic Sea, beyond where, to the west, my forefathers thought was the cosmic tree, the axis of the world, holding up the arch of the sky.’1
The Baltic languages are spoken today in Lithuania and Latvia, while the dominant languages in neighbouring Estonia and Finland belong to the Uralic family. Baltic wasn’t always spoken in those places, however. As with all languages, its territory has shifted over time. Gimbutas understood that the story of Baltic was intimately entwined with that of its neighbours in the region – Germanic, Slavic and Uralic – but she couldn’t help thinking of it as a case apart. In her mind, the Balts and their languages had stayed closer to their roots by virtue of their isolation in a land of forests and lakes, far from the highways of ancient Eurasia. She pictured her countryfolk doing the same things they had done for centuries, if not millennia: tilling the earth, living in harmony with the seasons, singing as they worked: ‘The Baltic area is exceptional among the lands inhabited by people of Indo-European origin in that the language and folklore have survived in a remarkably pure state …’
She wasn’t the only one to express this thought. It was received wisdom, in 1962, that Lithuanian in particular was a frozen vestige of Proto-Indo-European. The idea went back a long way. ‘Whole Sanskrit phrases are well understood by the peasants of the banks of Niemen,’ stated an 1882 edition of the Encyclopaedia Britannica, referring to the river that rises in Belarus and traverses Lithuania on its way to the Baltic Sea. ‘Anyone wishing to hear how Indo-Europeans spoke should go and listen to a Lithuanian peasant,’ wrote the French linguist Antoine Meillet in 1913. And it’s still often said, both inside and outside academia, that Lithuanian is the living Indo-European language that most closely resembles that long-dead ancestor. What was not to intrigue a young historical linguist searching for a specialisation in the late 2010s? But as soon as Anthony Jakob started looking under the hood of the Baltic languages, he realised that the truth was both more nuanced and more interesting.
Lithuanian sounds archaic, Jakob says, in part because of a quirk it shares with its extinct relatives, Ancient Greek, Latin and Sanskrit, which is that many of its nouns end in s in the singular. The Lithuanian word for ‘man’, výras, is very similar to its Sanskrit counterpart vīrás, while Lithuanian ugnìs (fire) resembles Sanskrit agnìs (‘man’ and ‘fire’ have become bīr and āg respectively in Hindi). But in other ways Lithuanian has evolved quite as much as any of its relatives. It has updated the grammar of Proto-Indo-European by, among other things, laying off a number of verb tenses. And its vocabulary has seen a comparable rate of turnover, accumulating substitutions as is normal in the process of language evolution. The Lithuanian word galva (head) is unrelated to its Proto-Indo-European counterpart *k’erh₂- (the root of Hindi sir, Persian sar and English cerebral), and although Lithuanian has a word, esti, that is related to French est and English is (the third person singular of the verb ‘to be’), Lithuanians no longer use it. They say yra instead, while Latvians say ir. Nobody knows where these two words come from.
In fact, Lithuanian adheres strictly to the rule that there is no such thing as a pure language. Its peculiar mix of archaisms and innovations reflects the periods of calm that it has known, as well as the periods of upheaval, and the other languages that it has brushed with along the way. Jakob, who defended his doctoral thesis in 2023, is one of the more recent recruits to the effort to piece that backstory together from the linguistic evidence. The result has been a sea change in thinking about the origins of Baltic and Slavic.
In Meillet’s day, and still in Gimbutas’ day, Baltic and Slavic were considered sufficiently different that they must always have been distinct branches of the Indo-European family. Now, armed with more information and finer resolution, most linguists acknowledge that the differences between them were racked up relatively recently. Baltic and Slavic were once the same language, they believe, and even after they split remained close for two thousand years or more. Events triggered by the fall of the Western Roman Empire caused Slavic to barrel away from its more conservative sister, but Baltic didn’t stay the same, and the linguistic legacy of that rapid divergence masked the long period of stability that preceded it. It was that recent divergence that had led Meillet and others astray.
The realisation that Lithuanian isn’t the fossil it was once thought to be has solved some mysteries but accentuated others. If it has followed its own path, why does it have so much in common with Sanskrit? The similarities are there in the other extant Baltic language, Latvian, and in the Slavic languages too, but they are most pronounced in Gimbutas’ mother tongue, and they extend from sound to vocabulary and even mythology. The divine twins are called Ashvins in Sanskrit and Ashvieniai in Lithuanian, and both names evoke horses.
A seemingly outlandish theory, that Balto-Slavic and Indo-Iranian were, in turn, once the same language, has been gaining currency of late. The theory isn’t new, but it has always been controversial, and it’s not hard to see why. An ancient nexus between languages that would eventually be spoken so far apart – one among the ruined metropolises of the Indus Valley, the other on the amber-strewn beaches of the Baltic – strikes many as beyond the bounds of probability. And yet in the last few years archaeologists and geneticists have produced evidence that such a nexus really existed. If you pull on the cultural and genetic cords that tie the speakers of Sanskrit and Lithuanian to their roots, you find that they cross in a single place and time: the Dnieper Rapids, in the Bronze Age.
By 2800 BCE, the people of the Corded Ware culture had already scorched a path into central and northern Europe. In the east they had gravitated like so many of their ancestors to the Middle Dnieper, south of modern Kyiv. There, where the water roared over a series of rocky shelves, they supplemented their diet of milk and meat with freshwater fish.
For hundreds of years, all their needs were satisfied in this land of plenty. Starting around 2400 BCE, however, the first augurs of the climate crisis would have made themselves felt, and as the colder, drier conditions intensified, and it became harder to eke out a living, some raised their eyes to the east. The migration that now got underway culminated in the Sintashta culture, the presumed speakers of Proto-Indo-Iranian who lived south-east of the Urals. But it began with the herder-fishers of the Middle Dnieper, who set off towards the Volga and the mountains beyond in search of fresh pasture for their bony herds.
The Sintashta were expert metalworkers and chariot-builders, but fundamentally they were herders. They formed the eastern extremity of a cultural continuum that began with the Corded Ware clans on the Dnieper. Genetically, too, the two groups were related. These findings have emboldened some linguists to claim that laid over the cultural and genetic continua was a linguistic one: a spectrum of dialects that bridged Balto-Slavic and Indo-Iranian. Eventually these would split into two branches that would go their separate ways (and each would then split again, and again), but for a while, people living throughout that wide space could understand each other. It’s a mind-bending thought, and most historical linguists do not (yet) subscribe to it. But a growing number do, and the reason is that a dialect continuum across the western steppe could explain those puzzling similarities between Balto-Slavic and Indo-Iranian.
Of the features that Balto-Slavic and Indo-Iranian share, the ones that linguists consider most revealing in terms of reconstructing the two branches’ past are satemisation and another sound change called the ‘ruki rule’. Under the ruki rule, an s inherited from Proto-Indo-European was systematically transformed into a sound resembling English sh after r, u, k or i. (An example is Proto-Indo-European *h2eusōs, meaning ‘dawn’, which became *austrōn- in Proto-Germanic – the root of English ‘Easter’ – but something closer to aushra and ushas in Lithuanian and Sanskrit respectively.)
These two sound changes affected both Balto-Slavic and Indo-Iranian (and in the case of satemisation, some other branches of the family as well), and they are thought to have taken place at roughly the same time. Some linguists consider the ruki rule to be so specific that – like some fluke genetic mutation – it can only have arisen once, and that Balto-Slavic and Indo-Iranian must therefore have been a single language when it happened. Others argue that it could have come about independently in two distinct branches. You can see why so much rides on this question, and why the ruki rule is so hotly debated. If Balto-Slavic and Indo-Iranian had to have been one language when they were ‘rukied’, Indo-Iranian must have crossed the steppe to reach the parts of Asia where its descendants are spoken today. If not, it could have left Anatolia with Neolithic farmers, passing south of the Caspian Sea and never coming within earshot of Balto-Slavic. On the ruki rule rests the origin story of the entire Indo-Iranian diaspora.
That there was once a nexus between Baltic and Slavic is far less controversial, given how long they have been neighbours, and is now widely accepted. Many of the tributaries of the Dnieper in its middle and upper reaches – in modern Ukraine, Belarus and Russia – have names that linguists deem to be Baltic or Balto-Slavic in origin.2 And Baltic and Slavic share a fairly rich body of vocabulary connected with rivers and fishing. Latvian and Lithuanian, spoken to the east of the River Niemen, once had a sister called Old Prussian that was spoken to the west of that river. The Baltic language Old Prussian had a word for ‘sturgeon’ (esketres) that is related to the name for the same fish in Russian, a Slavic language (osëtr). Lithuanian (Baltic) šãmas and Russian (Slavic) som spring from a single root meaning ‘catfish’. And Baltic and Slavic share words for ‘wading’, ‘diving’, ‘spawning’, ‘dugout canoe’ and ‘raft’ that have no counterparts in any other branches of the Indo-European family.
Linguists infer that Baltic and Slavic were fused at the time that they invented or borrowed this riverine vocabulary. They cannot have inherited it from Proto-Indo-European, because Proto-Indo-European lacked it. And since Baltic and Slavic have almost no agricultural vocabulary in common, the speakers of Balto-Slavic are unlikely to have been farmers. Their lexicon is a snug fit for the herder-fishers of the Corded Ware culture who archaeologists say were living along the middle stretch of the Dnieper in the early third millennium BCE.
Balto-Slavic probably remained a single language until about 2000 BCE, when some of its speakers started to move north and east, towards what is now Moscow and the upper reaches of the Volga. As discussed, there are scholars who think they were following in the footsteps of the Indo-Iranian speakers to whom they were also related genetically, but these later migrants spoke a forerunner of Baltic.3 The linguists who meticulously surveyed the river and place names of central and eastern Europe, in the early twentieth century, tentatively mapped the prehistoric range of the Balts as stretching from Poland to central Russia – as far east as the River Oka, a tributary of the Volga. On that basis, Gimbutas wrote, the Balts were misnamed. They had once lived far from the Baltic Sea, and before the expansion of the Slavs and Germani out of their respective homelands had inhabited an area six times the size of their modern territory. But she was writing before linguists recognised an initial Balto-Slavic phase of the language. It seems more likely now that Balts and Slavs named some of those rivers jointly, before they saw each other as Other. Slavic may have been the language that developed among those who stayed behind once the future Balts had left. Some of them followed the Dniester towards the Carpathian Mountains and present-day Poland, but eventually, as we’ll see, they would spread much further afield. The genetic evidence, where it is available, corroborates this scenario. The diversity of Y chromosomes among modern Slav men can be explained by a small, initial cluster that radiated out of the Middle Dnieper Basin.
Though they followed different trajectories, both Balts and Slavs now entered the forests that, four thousand years ago, covered all of northern and central Europe. In doing so, they encroached on farmers’ territory. From the villagers with whom they now entered into dialogue, they received their primary schooling in woodland living. They learned to haggle in beans, carrots and turnips, and to point out elks, woodpeckers and hawks to each other. We know that they had parted ways by the time this happened, because they borrowed different words for the same species. And it wasn’t just their vocabulary that had begun to diverge. Baltic and Slavic were obeying different sound laws by now. They had become distinct languages. They would remain close for a very long time, but they wouldn’t meet again for another two thousand years.
The Proto-Slavic homeland is rather fuzzily defined, especially in the west, but most scholars locate it roughly between the headwaters of the Dniester and the Middle Dnieper. It fell within the borders of modern Ukraine, in other words, close to the frontier with Poland. From the eighth century BCE, the Slavs’ neighbours on the prairies to the south and east of them were the brilliant, bejewelled Scythians, those Iranic-speaking heirs of the Sintashta who outsmarted Darius, for a time, and who managed to impose their own names on many of the region’s most significant waterways – that drumbeat of Ds.
In the early centuries of the Common Era, the East Germanic-speaking Goths came through the Slavic heartland, on their way from the Baltic island of Gotland to the Black Sea. A legacy of this meeting is the Russian word for a parliament, duma, which Gothic loaned to Slavic. (The original Gothic word was dōms, meaning ‘judgement’.) By the fifth century the Roman Empire had fallen in the West, and Europe had entered a period of turmoil which German-speakers call the Völkerwanderung (wandering of peoples). The climate cooled again. Wars and dangerous diseases ravaged the continent. It was a time of danger and opportunity, and the Slavs took to the highways.
They did so initially as part of a military federation ruled by Huns and Avars, Turkic-speaking horsemen from the steppes. Some went west, crossing the Elbe into German-speaking lands, but the majority followed the Goths south-east into the Balkans. If the stick driving them was the cold weather, the carrot luring them on may have been the lands vacated by Justinian’s Plague. There are souvenirs of these campaigns embedded in their language; postcards from the edge. From Old High German in the west, for example, they borrowed król, meaning ‘king’. This word, which is clearly unrelated to Latin rēx, Sanskrit rāj- or Gaulish rīx, is a corruption of the name of the Frankish king Kar(a)l, better known to English-speakers as Charlemagne.
The transformation of Slavic probably began with the attempts of Slavic combatants to emulate the speech of their Avar commanders. (The Avars might have lent the Slavs a word meaning ‘gathering place’, maidan in Ukrainian, that they in turn had borrowed from Iranic.) It continued as the people the Slavs conquered attempted to emulate them. One of the most important changes the language underwent, during this period of upheaval, was something that linguists refer to as the ‘open-syllable conspiracy’. An open syllable is one that ends in a vowel; a closed one ends in a consonant (think ‘go’ versus ‘god’). Most languages tolerate both, but during the Völkerwanderung the Slavic languages gradually ejected all closed syllables. In the earliest documented Slavic language, Old Church Slavonic, the word for ‘head’ that we met previously in its Lithuanian form, galva, became glava. A letter switch turned the closed syllable gal- into the open one gla-. Lithuanian ranka (hand) became ro˛ka, by folding the n into the nasalised vowel o˛, and Lithuanian ragas (horn) became rogu˘, simply by dropping the final consonant.
By a variety of tricks, in other words, Slavic tore itself wide open. And though that transformation began with Proto-Slavic, it continued even after the proto-language had begun to fragment into its daughters. Hence the label ‘conspiracy’, because it looks as if there had been a conscious and sustained effort to strip out every last closed syllable from Slavic. In reality, since it happened over many generations, and since people are usually unaware of how sounds are changing in their mouths, it cannot possibly have been conscious, and linguists have long puzzled over what caused it.
One theory, put forward by linguist Eugen Hill, has to do with the fact that all languages prohibit certain combinations of consonants. Modern English doesn’t allow dv, for instance, while Czech does (think of the composer Dvořák). Nor does English allow kn, today, but the spelling of words like ‘knife’ indicates that it did in the past – and that both letters were once pronounced. At the time of the Slav migrations, most of the Balkan peninsula apart from Greece was Italic-speaking – having previously been claimed by the Romans – but the Slavs would have encountered pockets of East Germanic-speakers too. Italic forbade the combination dl, then, while Germanic barred kt and pt. As the Slavs conquered speakers of these languages, the vanquished started speaking Slavic, but they stumbled over words containing consonant combinations that were prohibited in their mother tongues. (Psychologists report that people don’t even hear the combinations that are forbidden in their native language; they hear the nearest pairing that’s allowed, which they then reproduce.) The result was that Slavic absorbed prohibitions from a number of foreign languages which, combined, forced out its closed syllables.
Hill’s theory is pretty new, and at the time of writing his academic peers have yet to pass judgement on it, but there is some historical evidence to back it up. The Slavs are reported to have been remarkably open-minded towards those they subjugated, often accepting them as full and equal members of Slav society. This was unusual. The Anglo-Saxons, reaching Britain at about the same time, were more inclined to enslave the Celtic-speaking natives. But the Slavs’ openness to foreigners and their oddities of speech might have expedited the open-syllable conspiracy – or compromise. Slavic spun away from Baltic and simultaneously diverged three ways, forming western (including Polish and Czech), southern (such as Serbo-Croatian and Bulgarian) and eastern (Ukrainian, Russian and Belarusian) sub-branches.
By the end of the sixth century, according to Roman chroniclers, Slavs and Avars were busy laying siege to Thessaloniki, a port on the Aegean and the second most important city in the Eastern Roman Empire after Constantinople. The Thessalonians spoke mainly Greek, but when the city fell they learned to speak Slavic, probably as a second language to begin with. It was therefore not so strange when, receiving a request from a Moravian prince to send missionaries to spread Christ’s teachings among his people in the ninth century, the Emperor and the Patriarch of Constantinople looked for candidates among the Thessalonians. The patriarch selected two brothers, Constantine and Michael, for the mission to enlighten Slavs in what are now the Czech Republic and Slovakia.
Though the brothers spoke Slavic, all religious texts existed in other languages at the time, and an important part of their mission was to translate them. The problem they faced was that the Greek and Roman alphabets were not well suited to Slavic, meaning that they failed to capture some sounds while including others that Slavic didn’t use. Constantine, the younger and more intellectual of the brothers, set about adapting the Greek alphabet so that he could more accurately transcribe the Slavic dialect that he and Michael had grown up with. Since the first documents to record Slavic were the brothers’ Bible translations, that dialect, long dead except as a liturgical language, is known as Old Church Slavonic. The brothers are better known as Saints Cyril and Methodius, and the alphabet that Constantine devised would evolve into Cyrillic. (Roman Catholicism had replaced the Orthodox Church in Moravia before the ninth century was out, along with the Roman alphabet, but the Cyrillic script came to dominate further east.)
The western and southern branches of Slavic lost touch after the arrival of the Magyars from the late ninth century drove a Uralic-speaking wedge between them, in Hungary. The expansion of an Italic language, Romanian, helped push apart the southern and eastern branches, and in time the three Slavic-speaking populations ceased to be able to understand each other. As for their shared name, it’s fairly common knowledge that ‘Slav’ is linked to English ‘slave’, but that obviously wasn’t its original meaning (what people would boast of being ‘the unfree’?). Some linguists see in ‘Slav’ a derivation of Proto-Indo-European *k´léwos, meaning ‘fame’, in which case Slavs may have called themselves something like ‘the famous people’.
Starting in the Middle Ages, when Franks took Slavs captive at the borders of the Holy Roman Empire, the Slav name – sclavus in the official language of that empire, Latin – became associated with servitude. (You may recall that Latin already had a word for a domestic slave, famulus; new times and new categories called for new vocabulary.) Latin sclavus morphed into French esclave, which became ‘slave’ when the Normans took it to Britain. But the Slavs themselves retained the first meaning of the word, as witnessed by the Russian boy’s name Mstislav (vengeful fame), the Polish one Stanisław (he who has achieved fame), and the Ukrainian war cry Slava Ukraini! (Glory to Ukraine!)
At some point in the Balts’ migration towards the eponymous sea, from what is now Ukraine, a group broke away and headed for lands west of the Niemen (roughly the modern Russian exclave of Kaliningrad). Their language became Old Prussian which, despite the misleading inclusion of ‘Prussia’, was a Baltic and not a Germanic tongue. The remaining Baltic-speakers eventually reached the forests north of modern Moscow, where they encountered barbarian hunters sporting sumptuous furs.
The barbarians’ language was Uralic – Proto-Finnic, to be precise, the ancestor of modern Finnish and Estonian.4 They were the bow wave of the western expansion of the Uralic languages out of their homeland, which is now thought to have been located – not in the Urals as their name would suggest – but closer to Siberia. (Proto-Saami, a sister of Proto-Finnic, had already departed towards Lapland, where it would become the language of reindeer-herders.) The meeting of the Proto-Balts and the Proto-Finns unfolded in an unexpected way.
Fortified villages began to spread across the region east of the Baltic Sea, as far north as Finland and as far east as Moscow. You can see their remains today, jutting out of riverbanks and lakeshores. The inhabitants of these forts obviously feared someone; they built ramparts of stone or baked clay several metres high, and reinforced them with timber to keep the enemy out. But the enemy may just have been the inhabitants of the next fort along. It wasn’t that the Balts feared the Finns or vice versa, because the two groups cohabited peacefully within.
Distinct pottery types and burial rites have been found in the settlements, which suggest to Estonian archaeologist Valter Lang that Balts and Finns found a way of living together while maintaining their ethnic and cultural differences. Linguists, too, suspect that they formed mixed bilingual communities in which they preserved their mother tongues but loaned words to each other. The Balts were fully fledged farmers by now. The Finns were forest specialists who hunted game and brought back valuable skins and furs. Through a process of trial and error, they may have discovered that they were more prosperous together than apart. The Finnish words for ‘seed’, ‘pea’ and ‘chaff’ are all Baltic loans (Lithuanian/Finnish sėmuõ/siemen, žir~nis/herne and pe~lūs/pelut respectively). And during this period of cohabitation, both languages absorbed loans from an unknown third. Since many of the loans relate to fishing – ‘fish weir’, ‘crayfish’, ‘whitefish’, ‘eel’ – the source may have been a language spoken by hunter-fishers who had inhabited the region long before either Balts or Finns arrived.
How close you are
My forefathers!
They herded cows
And saddled horses,
Planted children
And peas.5
The Balts and Finns had been living this communal life for more than a thousand years when the Slavs showed up in their midst. Their language had once been indistinguishable from Baltic, and in terms of vocabulary and to some extent grammar the two were still close. But thanks in large part to the open-syllable conspiracy, Slavic now sounded quite different from its sister. The Slavs’ wanderings, in which the Balts had not participated, would have been audible. They settled among the Balts and Finns, and by the ninth century speakers of all three languages – two Indo-European and one Uralic – were mingling in the north-western Russian city of Novgorod, close to the borders with modern Finland and Estonia.
Novgorod was a strategic hub on the river network that connected the Baltic south via the Dnieper to the Black Sea and Constantinople, and east via the Volga to the Caspian and Baghdad, and the Slavs weren’t the only ones drawn to it. Vikings had crossed the Baltic from Sweden at about the same time, and established a trading colony there. The Slavs may have carved out some indispensable role for themselves in the city’s commercial life, since the language of trade there was an East Slavic dialect that linguists call Old Novgorod. We know this because the city’s inhabitants wrote to each other (some of their letters, scrawled in Cyrillic on birch bark, are on display in the State Historical Museum in Moscow). Their correspondence is the first documentary evidence of a Slavic vernacular, since the older written language, Old Church Slavonic, was mainly a language of nobles and priests. But the letters are also precious because the oldest of them, dating to the eleventh century CE, were written before Balts and Finns had started to record their own languages, so they provide a glimpse of a city in whose streets literate and preliterate people mingled. We can be confident that the Balts were present, taking their cut of the lucrative sale of squirrel pelts, because the Baltic language Lithuanian borrowed a Slavic word for a now-obsolete unit of weight known in English as a ‘ship-pound’ (bìrkavas in Lithuanian, from Old Novgorod bérkovec).
The birch-bark letters speak to a highly literate population, since among the authors were women and children. Besides endless inventories of furs they include a marriage proposal, a dispute over a stolen female slave, and a little boy’s drawing of himself as a brave warrior. The authors cursed, cracked jokes, spread scurrilous rumours and signed off with formulae such as ‘I kiss you’ or ‘be so kind’. Happily for historical linguists, one fourteenth-century missive included a small Finnic–Slavic dictionary. The city in which the letters were composed was heavily fortified, with log-lined roads, and stone churches and monasteries that were treasure houses of icons and frescoes. Medieval Novgorod was known for its art, a reputation it maintained after coming under Muscovite rule in the fifteenth century. But it was that change of regime that ushered in a new East Slavic dialect, a forerunner of Modern Russian, and caused Old Novgorod to fall into disuse. At about the same time, Balts and Finns finally took up writing.
The Vikings, or Varangians as they were known in the east (probably from the Old Norse word varing, meaning ‘ally’), would have spoken Old Norse to begin with, but like Viking invaders everywhere they took up the language of the people they conquered.6 These formidable mercenaries in their clanking chainmail were among the chief enslavers of Slavs, yet they embraced the local Slavic tongue so quickly – within a couple of generations – that they barely had time to leave their mark in it. Their arrival in the east is nevertheless immortalised in the name ‘Russia’, since they were known in Novgorod as Rus (sometimes written Rus’, to indicate a soft s sound at the end). Linguists don’t agree on where the name Rus came from, but one theory traces it to the Finns’ name for Sweden, Ruotsi, whose literal meaning might have been ‘oarsmen’.
In time the Varangians set up trading posts further south, notably at Smolensk and Kyiv. At the Dnieper Rapids they hauled their flat-bottomed boats out of the river and rolled them over logs until it was safe to put them back in the water. Influences flowed upstream too, from Constantinople, and in 988 CE a prince of Varangian heritage named Vladímir converted to Orthodox Christianity. Vladímir, which means ‘conquer the world’, is said to have had his entire court baptised in the Dnieper. The religion spread, and over the next few centuries Rus came to refer to all Orthodox Eastern Slavs. Only later did the name become attached to a geographical space, and later still was that space divided into Russia, Belarus and Ukraine. The Finns were Catholic, having been converted by the Swedes, but in Novgorod Eastern and Western Churches met, and it was here, from the Orthodox Slavs, that the Finns took their words for ‘priest’ (pappi) and ‘cross’ (risti). The Balts breathed the incense but rejected the creed. They remained stubbornly pagan, continuing to sacrifice horses, in rituals resembling the ashvamedha, right up until the thirteenth century. They even imported those horses, by boat across the Baltic, from their Christianised neighbours in Scandinavia.
The first Balts to convert were the Old Prussians, who had no say in the matter. The invasion of the crusading Teutonic Knights marked the beginning of the centuries-long decline of their language, as German-, Polish- and Lithuanian-speakers moved into their territory. The last speakers of Old Prussian probably died of famine and plague around 1700. The Latvians were next to embrace the Western Church, but the Lithuanians continued to hold out. After decades of vacillating between Eastern and Western options, Europe’s last pagan nation capitulated in 1387, joining the Western Church.
It did so for reasons of political expediency. In order to marry the young Queen Jadwiga of Poland and claim the Polish throne, Jogaila, Lithuania’s pagan Grand Duke, had to agree to convert himself and his country to Catholicism (he took the baptismal name Władysław, meaning something like ‘famous ruler’, and the dynasty he and Jadwiga founded became known as Jagiellon). The royal union ushered in four centuries of shared Polish–Lithuanian history, the last two of which saw the two countries form a single state. Balts and Slavs were united again, as they had been four millennia earlier, but not for long. Austria, Russia and Prussia divided the state up between them, and by 1800 the Polish–Lithuanian Commonwealth had ceased to exist.
The next time you find yourself in Kraków, Poland, go to the Wawel, the traditional seat of Polish kings, and head for the cathedral.† Just before you pass through the great door into the realm of dead royalty and some exceedingly fine stained glass, look up. Suspended by chains above your head is a jumble of monster bones. If they seem out of place, if they strike you as pagan, you are right on both counts. They are the bones of a whale, a mammoth and a rhinoceros that were pulled from the silt of the River Vistula (Wisła to the Poles) when the foundations of the cathedral were being laid in the eleventh century CE, though legend assigns them to a dragon called Smok.
A short walk across the Wawel complex, a hundred and thirty-five steps lead down in a spiral to Smok’s supposed lair. You have to buy a ticket to enter, but if you ask nicely, the person who takes it from you will tell you the legend: how Smok terrorised the land, devouring livestock (or in some versions, maidens); how the sons of King Krak defeated the dragon, but then one son killed the other and was banished, so the kingdom fell to their wise and fair sister Wanda. Emerging from the cave, you find yourself on the bank of the Vistula, where a green-tinged statue of Smok breathes actual fire.
Smok or Żmij, or Zmei or Zmaj, depending on which Slavic-speaking country you happen to be in, is the archetypal serpent, denier-of-life, and any resemblance you may notice to J. R. R. Tolkien’s dragon Smaug is not coincidental. Tolkien was a philologist who famously created The Lord of the Rings as a vehicle for languages that he had constructed, and he chose that name for its (to him) pleasing Indo-European ring. He knew that a natural language is always associated with a culture, and that every culture has its mythology. He therefore invented the legends that would make his languages live. (Tolkien referred to Esperanto and other constructed languages as ‘dead, far deader than ancient unused languages, because their authors never invented any Esperanto legends’.7) There was actually a Proto-Indo-European word, *smeuk, that probably meant ‘to slide’ or ‘glide’, and if the Slavic dragon names are derived from it then they are living exhibits of taboo deformation – the phenomenon whereby taboo words are rapidly recycled through euphemism and circumlocution. (In this case, a word describing the creature’s motion may have come to replace its unspeakable name.)
During the communist period, Poles compared themselves to radishes: red on the outside, white on the inside – red and white being the colours of the Polish flag. The implication was that their communism was only skin-deep; beneath it they were Catholic, and capitalist. But beneath the Catholic layer is yet another: the pagan. The construction of the Wawel cathedral began just decades after Poland became officially Catholic, in 966 CE, and while it was being built a pagan rebellion got underway that nearly destroyed the nascent Polish state. Smok’s bones serve as a reminder of that older layer, and of that near-annihilation.
In many Indo-European traditions the life-negating serpent became a milk thief, a ‘cow-suckler’ that wasn’t past nestling in among the hungry calves, cuckoo-like, to claim its share. Biologically speaking this is impossible – snakes can’t suck, and they lack the enzyme that would allow them to digest mammalian milk – but the idea is firmly cocooned in the Indo-European mind. A Hindi word for a reptile, godhā, literally means ‘cow-suckler’ – from go, ‘cow’, and dha, ‘to suckle’ – while Baltic names for a grass snake (Lithuanian žaltys, Latvian zalktis) are possibly derived from Proto-Indo-European ones meaning ‘delighting in milk’. Here, again, the ancient nexus of Lithuanian and Sanskrit may reveal itself. But the Balts developed the cow-suckler motif in their own way, in their land of lakes and forests. They alone turned the snake into a friend, paying it in milk to protect the farm. If it wasn’t fed it would unleash dark forces, but as long as you gave it what it wanted it would bring you luck. In pagan times, Baltic priestesses fed milk to snakes as part of their fertility rituals. ‘It was a blessing to have a žaltys in one’s home, under the bed or in some corner, or even in a place of honour at the table,’ wrote Gimbutas in The Balts.
Perhaps it was inevitable, then, that the grass snake would become the symbol of pagan resistance to Christianity. Medieval chroniclers relate that the German crusaders seized the snakes from the village houses and threw them on to bonfires. After the Lithuanians converted in the fourteenth century, crosses began to spring up all over the Lithuanian countryside, but often they had snakes carved into them. Depending on your point of view, the snakes were Christianised, or the crosses were paganised. And although the Lithuanian language is far from the throwback it was once thought to be, the country’s pagan past still lies close to the surface. Well into the twentieth century it was not uncommon to hear a Lithuanian say, Kur žalčiai yra, tai ten tie namai yra česlyvi. ‘Wherever grass snakes are, the house is full of happiness.’
† Wawel is pronounced Vavel in English, while the girl’s name Wanda is pronounced Vanda. As an ad campaign for a popular brand of Polish vodka made clear in the noughties, there’s no v in wodka. In the Polish alphabet the letter ł is pronounced like English w, while the letter w is pronounced like English v.