Why do some ancient languages become widespread across cultures, while others don't?

Question

At first glance, I don't see any strong interconnections between the spread of languages and the history of conquests.

For example, according to the Northern theory, during the Middle Ages, Eastern Slavic tribes were conquered by the Varangians (Vikings). The Primary Chronicles state that the first rulers of the Kievan Rus were Varangians, and I assume that the majority of the authorities were as well. However, the ancient Germanic language didn't become the basis for modern Russian. Modern Russian clearly belongs to the Slavic group of languages, perhaps with some minor Germanic borrowings at best. Therefore, it appears that the initial Germanic tribes were completely assimilated by the conquered Slavs, both culturally and genetically.

An opposite example is the conquer of Anatolia by the Turkic people. Some genomic research shows a small genetic impact of the ancient Turkic people on the modern population of Turkey, yet the modern Turkish language belongs to the Turkic language group.

Delving deeper into history, what intrigues me is how the native language of an Indo-European people, who initially lived in a small area of Eurasia, became so widespread across the continent. This is especially fascinating considering that modern people who speak distinct "dialects" of the Proto-Indo-European language have little to no genetic connections to the Indo-European people.

I assume that my question has close relations to linguistics and archaeology as well, but I would like to know if there are any general historical science theories about the widespread nature of languages.

Answer 1

The special case: rulers assimilate into the community

The Varangians are a drastically different case from the Turks, and just about every other scenario I can think of. According to our only source they were invited to rule by the Slavic peoples, and they arrived with a retinue of warriors rather than a group of settlers. Rurik then spread those men across a large number of cities:

И принял всю власть один Рюрик, и стал раздавать мужам своим города — тому Полоцк, этому Ростов, другому Белоозеро.

And Rurik took power, and distributed cities to his men - Polotsk to one, Rostov to another, Beloozero to yet another.

While the Varangians were acting as governors, they had neither the mandate nor the resources (nor possibly even the desire) to convert the locals to a Scandinavian culture

Варяги в этих городах — находники, а коренное население в Новгороде — словене, в Полоцке — кривичи, в Ростове — меря, в Белоозере — весь, в Муроме — мурома.

Varangians in these cities were strangers, and the people in Novgorod were slovenes, in Polotsk - kriviches, in Rostov - meryans, in Beloozero - wes, in Murom - muromians.

These peoples were not only living in their own separate towns, but part of a wide-ranging culture stretching (depending on how you draw boundaries) to the Black Sea. The appeal to start speaking and acting in a Scandinavian manner would have been nil.

The Varangians reigned for an extremely brief period of time - just 80 years after Rurik, we have a ruler who appears to be culturally Slavic in Svyatoslav, and possibly even before him in Olga. It is no surprise that they failed to make a significant impact on the language.

The general case: settlers bring the language with them

The Turks didn't just show up with a couple of boats full of guys - they had a sustained migration westward from the Eurasian Steppe. The process of Turkic settlement in Anatolia happened over generations, alongside Byzantium's decline in the area. Anatolia was contiguous with a massive and wealthy Turco-Persian empire, so Turkish rulers could (and did) Turkify the population, a term that even appears in Greek in the 1300s as 'εκτουρκισμός'.

Nothing in history is cut-and-dry

Between these extremes, there are a lot of intermediate examples. Consider the Manchu rulers of China. While the Empire continued to speak various East Asian languages (China was far too massive to enforce any single language on the populace even back then), the Emperors tried to maintain their Manchu culture. 300 years after the Manchu first conquered the throne, the last Emperor could not speak Manchu, but still studied it.

We can also look at English. Successive invasions (Anglo-Saxon, then Danish, then Norman) layered new words onto the languages spoken in the British Isles until the original Brittonic languages were eclipsed. But English was not the language of William, or Ragnar Lodbrok, or any of the other invaders and conquerors.

Answer 2

My main principle in looking at these things is that demographics wins, and farming > herding > hunting & gathering.

So when you're looking at which language won out in an area, you're largely looking at who ended up doing the farming there.

Often, particularly in the early Neolithic, such as the Indo-European takeover of Europe, what you see is largely that someone who had farming moved into an area where any existing residents didn't, or didn't know how to do it as well.

Sometimes instead what you see is some combination of societal collapse and ethnic cleansing. This is what happened with Germanic Anglo-Saxons moving into England and replacing the Celtic languages spoken on that island except in the margins.

Sometimes instead what you see is a ruling class that is content with just being that, and letting the original farming peoples continue to handle the drudgery of farming. This is what happened with the Norman Conquest.

So you can apply this principle anywhere you see two languages clash, and one eventually win out. Varangians? Content to be a ruling class. Franks and Goths? Content to be a ruling class. Romans in the same area? Ethnic cleansing. Turks? See SPavel's answer.

As for the situation with genetic studies in Turkey, the findings there are quite varied. Different studies have found different things. Halpotype-wise, I'm seeing where they actually seem to share a lot of halpotypes with other Turkish-speaking groups, like Uzbeks.

There are also many ways genetic material can flow around. Its no respecter of political, cultural, or linguistic boundaries, particularly in an area that sees a lot of trade and is a natural geographic crossroads like Anatolia. Due to natural selection, genes are sometimes encouraged to move or grow where any attached culture might not be.

Answer 3

I assume that my question has close relations to linguistics and archaeology as well, but I would like to know if there are any general historical science theories about the widespread nature of languages.

This final paragraph is intriguing. I am not sure if historiography can be written without contribution from linguistics and archaeology. That would be ignoring important data of human history.

On MCW's reference (comments and answer, above) to "The Horse, The Wheel and Language" by David Anthony, this book is important precisely because it incorporates historical linguistics. David Anthony's defence of historical linguistics as a key element of archaeology, pp. 16-7, is a reminder of its relevance to historiography.

If we accept historical linguistics as part of "general historical science", the most persuasive answer would be "Farming-Language Dispersal hypothesis" (Wikipedia). Simply, population density from agriculture provided excellent conditions for prehistoric societies to disperse.

The Encyclopedia of Global Archaeology (2nd edition, 2020), also tells us that historical linguistics, although overlooked often, it should not. I suppose I am doing the same, in this answer. The entry at page 4176, on Farming-Language Dispersals: Principles:

Among various potential "window on our past," once rich seam of data is all too often overlooked: the linguistic record. Above all, much of the world so so dominated by just a handful of vast language lineages as to cry out for explanation. That explanation, moreover, can only lie in the same contexts and processes that humanity's cultural and population (pre)history more widely. Indeed, of all attempt to account for these broadest patterns in the linguistic panorama, one of the most ambitious generalization is the "farming-language dispersals hypothesis." Most simply put, this proposes that many of the most significant language families - in both geographical range and speaker numbers - dispersed with, and primarily thanks to, the spread of agriculture.

---

In the case of Indo-European languages, it was genetic adaptation to milk tolerance of adult Yamnaya society, around 4000 BCE, that allowed the prehistoric migrations of Proto-Indo Europeans.

From the abstract of a recent article, Wilkin, S., Ventresca Miller, A., Fernandes, R. et al. Dairying enabled Early Bronze Age Yamnaya steppe expansions. Nature 598, 629–633 (2021) - https://doi.org/10.1038/s41586-021-03798-4

The rapid onset of ubiquitous dairying at a point in time when steppe populations are known to have begun dispersing offers critical insight into a key catalyst of steppe mobility. The identification of horse milk proteins also indicates horse domestication by the Early Bronze Age, which provides support for its role in steppe dispersals. Our results point to a potential epicentre for horse domestication in the Pontic–Caspian steppe by the third millennium bc, and offer strong support for the notion that the novel exploitation of secondary animal products was a key driver of the expansions of Eurasian steppe pastoralists by the Early Bronze Age.

---

Finally, having said the above, I believe it should be more than a single-factor (agriculture or by-product of domesticated animals) that created the necessary conditions for successful disperses of so many ancestral languages. There has to be more factors that we have yet to identify. For instance, environmental conditions, in whatever form, could also have been a major contributor in some disperses. In this, I think I am supported by a very thorough and relevant book on this very topic, "Language Dispersal Beyond Farming" by Martine Robbeets & Alexander Savelyev (Max Planck Institute for the Science of Human History, Jena, 2017) - open access.

Answer 4

Couple of other comments.

1. The question seems to be predicated on survivorship bias. We know some cultures (see point three below) eventually displace other cultures. This is observed fact. But that obscures two truths. First, that we have more data about the cultures that survive than we do about the cultures that don't. And second, factors that in retrospect were important in the "victory" of the surviving cultures were not obvious at the time, and may have been important as a result of pure chance. If I hold a winning poker hand, what are the factors that caused it to be mine? In a neighboring reality, this might be a losing hand; the factors that determine the winning hand include both the cards in my hand and the cards missing from my competitors' hands. There are factors (my decision to play, etc.), but the dominant factor is random chance. In retrospect, the outcome is clear and observable. But analysis of the factors may have little or no predictive power.

2. I largely agree with the other answers, but would add that history is not quite so linear. The Horse, The Wheel, and Language points out that sometimes nomadic herders dominate settled societies and sometimes the opposite. The Mongol empire was far more successful than any of the other competing states. Until it wasn't. This is another idea that is ancillary to the book so not developed. If I were to pursue it with more research, I suspect that nomadic societies can, in general, dominate settled farmers, but in times of adversity, the higher birthrate of settled societies causes them to win out in the long run. Proto-Indo-European won out in the end, but there was nothing intrinsic about the PIE culture at start which caused that win.

3. I suspect that the general trend is that societies that can more effectively marshal power towards a common purpose are long term more successful - but only over the long term. Chingis Khan was effectively undefeated, but the successor states declined relatively quickly. Fukuyama's measurements of the effective state seem to apply - states that are more effective will dominate states that are less so. Unfortunately, there is a very weak tie to language groups.

4. The thesis of the book is peripheral to the question you ask - which also makes it difficult to quote.

5. One of the points of the book is that the notion of "culture" as a discrete entity is mostly a fiction that we create to simplify study. Looking back. Members of the Eastern Slavic tribes might not recognize the existence of such a culture group. Much like in colonial America - the notion that there were "united states" was a minority position. It wasn't until after the Civil War that one could positively state that the nation was stronger than the states (and that is in a post-nation-state era). Many cultures can share one language group (which complicates the question under discussion). As I understand it, it is relatively uncommon for one one culture to contain many language groups (which leads to the "stronger societies dominate" guide above). Might also be worth researching that Anthony claims that one of the few relatively reliable ways to identify a cultural group is at a self-defined frontier - when two cultural groups define themselves in opposition. This is likely to strongly influence the question you ask, but ignores the (probably far more numerous) cases where cultural and linguistic groups are not distinct. The first several chapters of the book address whether or not the PIE culture actually ever existed, or if it is an academic simplification

6. I'm listening to Horse/wheel/language as an audiobook, which makes it difficult to cite.

Having written too much on why it is difficult I'll nevertheless offer some speculations (hypothesis that I'd explore). First, societies that can harness & coordinate more effort are going to be more successful. Larger groups and more coordinated groups with more effective control will dominate smaller groups and those where control is diffuse or difficult. Second, resilience & flexibility will be key. History shows that even a strong culture can lose if the fundamental conditions change. ( I believe the Hydraulic Empire thesis has been discredited, but may be useful as thought experiment. Horse/Wheel/Language strongly suggests that one of the cultural horizons was likely caused by climactic or environmental change; those groups who had more options were more likely to survive. And absolute speculation - at the Yamnaya horizon, there is an increase in commercial activity and (as I understand it) new words added to the language for guest/host. Clearly conflict was going on at the same time, but my hypothesis is that the group that is capable of more complex social interactions with the "other" is more likely to survive. The group that has the option for commercial activity, assimilation, diplomacy as well as low and high intensity conflict will probably surpass those who have fewer options.