Agriculture and the Indo-Europeans – Steppe and South Asia

Proto-Indo-Europeans were farmers and not pastoralists

It is interesting to note that while a couple of decades ago and perhaps more, agriculture was not considered a part of Proto-Indo-European culture, it is now no longer the case. It was mistakenly believed that the Indo-Iranians must not have practiced agriculture because apparently the Indo-Iranians did not share an agricultural vocabulary with the rest of the Indo-Europeans (i.e. the European IE). However, more recent research has clearly shown this to be an error and it is now well accepted that the Indo-Iranians shared quite a significant amount of agricultural vocabulary with the other Indo-Europeans, sufficient enough to posit agriculture at the Proto-Indo-European stage.

The overall pattern of agricultural terms has been a persistent topic in IE studies, much of which has been stimulated by the observation that while stockbreeding terms appear to be widespread across the entire range of IE stocks, agricultural terms tend to be confined more closely among the European stocks and are, from a traditional point of view at least, scarce in the Indo-Iranian languages…

However,

…there is no case whatsoever for assuming that the ancestors of all the Indo-European stocks did not know cereal agriculture. While there may have been speculation in the past as to whether some terms might have applied originally to the gathering and processing of wild plants, terms for the plow, cultivated field, and techniques appropriate to the processing of domesticated cereals whose home range lay outside of most of Europe, suggest that all the earliest Indo-Europeans knew agriculture before their dispersals. (Encyclopedia of Indo-European Culture. Mallory & Adams, 1997).

The term that is cited as the major culprit for this error denotes a cultivated field in European languages while in Sanskrit it simply means a plain.

The second term (*haegˆros) has caused much discussion as the European cognates indicate a cultivated field (e.g. Lat ager, OE æcer [> NE acre], Grk agro´s, Arm art, all ‘field’) while the Skt a´jra- means simply ‘plain’ with no indication of agriculture. This divergence of meaning led to the proposal that the Indo-Iranians separated from the Europeans before they had gained agriculture so that we might posit a pastoral Indo-Iranian world and an agricultural European. Such a distinction is not borne out by the abundant evidence that Indo-Iranians also shared in an agricultural vocabulary… (The Oxford Introduction to Proto-Indo-European and the Proto-Indo-European World. Mallory & Adams, 2006).

…the Proto-Indo-Europeans possessed an economy based on domesticated livestock and domestic cereals. Earlier models such as those developed in detail by Wilhelm Brandenstein (1936) that suggested a marked dichotomy between arable Europeans and pastoral Indo-Iranians (or Tokharians) cannot really be sustained (Mallory 1997b) and despite a considerable number of differences there is still a substantial amount of shared agricultural vocabulary between European and Asian languages (Table 1 and 2). While the lists of cognates can certainly be criticized in certain specifics and they may well be an over-optimistic summary, I fear that there would still be a sufficient assemblage of words to indicate that both Europeans and Asiatic Indo-Europeans shared inherited words for both livestock and arable agriculture… (Twenty-first century clouds over Indo-European Homelands. Mallory 2012).

A brief overview of the shared agricultural vocabulary which indicates the state of agriculture among the proto-Indo-Europeans is as under –

Terms associated in general with agriculture can be found in both the European and Asian languages. Without specifying the precise nature of the cereal, we have at least six words for grains…The word *ga/ondh- “wheat?”, although cited in some handbooks because of its attestation in Hittite, Indo-Iranian, and Tocharian, is most likely an Asiatic loanword (Witzel 2006:97–99). There is also a word for “weed” or “rye” (*h2éreh2-) and an esculent root (*ālu-). There are also several words for the anatomy of a cereal or products of its processing, e.g., awn (*h2ekstí-) and chaff (*pelo/eh2-). The activities associated with plowing (*h2erh3ye/o-, *ghel-), harrowing (*h3ekéteh2-), and the hoe (*mat-) are preserved as well as all the major processes from sowing (*seh1-) to threshing (*wers-) and grinding (*melh2-, *peis-, *h2el-) as well as the word for sickle (*s r˳ po/eh2-).

While the exploitation of plants in the Asian IE languages seems secure, the specific nature of those plants is not. One of the most widespread words for a cereal is *yéwos and although it does designate “wheat” in Greek, it means “barley” in Armenian and Indic and either “barley” or simply “grain” in Anatolian and Iranian. Another word, *dr˳hxweh2- means “wheat” in Germanic but “rye” in Celtic, just “grain” in Tocharian, and it is a “grass” in Indic. “Barley” is the better bet for *yéwos and it is the only meaning we have for *h2élbhit, which is found in Greek, Armenian, and Iranian… (Indo-European Dispersals and the Eurasian Steppe, J P Mallory 2014).

The data shows that the early Indo-Europeans, even before the separation and dispersal of the Anatolian branch were familiar with the cultivation of wheat and barley and a few other cereals.

The reconstructed vocabulary for domesticated plants forms a restricted part of the botanical vocabulary as a whole although it is clear from the approximately twenty lexical items that the Proto-Indo-European community was familiar with cereal agriculture, particularly with wheat and barley, and there are at least half a dozen strongly reconstructed terms associated with planting, harvesting, and processing cereal grains. While this has little geographical importance it does indicate that Proto-Indo-Europeans must have had at least a Neolithic subsistence base, i.e. date no earlier than c. 8000 bc, and that there is no question of their adhering to some form of (largely mythic) pure pastoral economy. (Mallory & Adams, 2006).

 

No evidence of farming on the Steppe

The foregoing discussion makes it clear that there should be evidence of barley and wheat cultivation in any proposed Proto-Indo-European homeland as well as plentiful evidence of agricultural practices. However, this requirement puts a formidable obstacle for the proposal of the Steppe as the PIE homeland.

Mallory (2014) points out what ails the steppe…

We have so far seen that there is no certain evidence for the consumption of domesticated cereals in the steppe populations of the Eneolithic and Early Bronze Age and where there is some evidence for utilizing plants in the diet, e.g., among Yamnaya burials of the Caspian steppe (Shishlina 2008:232–33), all plant/pollen pollen remains identified belonged to wild plants such as chenopodium, artemisia, and ephedra.

So there is no evidence of agriculture among the Yamnaya folks who apparently brought the IE languages to the early European farmers.

But the problem is even more acute when it comes to the origin and migration of Asiatic branches (Indo-Iranian & Tocharian) of the Indo-European family from their proposed steppe homeland. Mallory (2012) citing none other than Anthony himself notes,

As Anthony remarks in this symposium, there is really no serious evidence for arable agriculture (domestic cereals) east of the Dnieper until after c 2000 BCE (see also Ryabogina & Ivanov 2011; Mallory, in press:a). This means that there is also no evidence for domestic cereals in the Asiatic steppe until the Late Bronze Age (Andronovo etc). From the perspective of the Pontic-Caspian model, the ancestors of the Indo-Iranians and Tokharians should not cross the Ural before c 2000 BCE at the very earliest.

The Sintashta culture is presumed to be the staging ground from where the Indo-Iranians marched southwards into the Iranian plateau and the Indian subcontinent. Yet the data from Sintashta settlements is equally damning,

Previous archaeological and archaeozoological investigations have revealed that the Sintashta economy was mainly based on livestock herding of cattle, small cattle and horses. There is additional evidence for fishing as well as some copper metallurgy, while the role of agriculture is still being discussed. Archaeobotanical studies therefore focused on the question whether the Bronze Age settlers cultivated, processed or stored crop plants. All cultural layers and important settlement features were systematically sampled, wet-sieved and subjected to plant macro-remain analysis. Anthracological (charcoal) samples were taken from burnt structures and charcoal concentrations. As no evidence for either cultivated plants or agricultural practices was found in the settlement, it can be concluded that farming was not practised at Kamennyi Ambar, and that the Sintashta economy was exclusively based on animal husbandry and fishing in that steppe region…

…From all the evidence, agriculture was not a part of the Sintashta economy, since neither cereals nor plants that could be related to any other cultivation practice were found. This supports recent research opinions stating that agriculture was not practiced in the Trans-Urals and adjacent regions until the Late Bronze or early Iron Ages (Epimakhov 2010; Anthony et al. 2005; Anthony and Brown 2007; Ryabogina and Ivanov 2011). (source : Archaeobotanical analysis of plant use at Kamennyi Ambar…, Ruhl et al. 2014).

Let me quote from another more recent study (Hanks et al. 2018),

The isotopic results from KA-5, and recent botanical and archaeological studies from the Kamennyi Ambar settlement, have not produced any evidence for the production or use of domesticated cereals. While this does not definitively answer the question as to whether Sintashta populations engaged in agriculture and/or utilized agricultural products, it does call into serious question the ubiquity of such practices across the region and correlates well with recent archaeological, bioarchaeological, and isotopic studies of human and animal remains from the Southwestern Urals region and Samara Basin. (Bronze Age diet and economy : New stable isotope data from the Central Eurasian steppes (2100-1700 BC))

In other words, neither during the Yamnaya phase nor doing the supposed Indo-Iranian Sintashta phase, there is any evidence of farming on the steppe, all the way from Dnieper to the Altai and whatever evidence there is only indicates the use of wild plants. This is strikingly at odds with the extensive agricultural vocabulary of the Indo-Europeans and the Indo-Iranians. What this effectively means for the proponents of the steppe theory is apparently the Indo-Iranians somehow miraculously preserved agricultural vocabulary from the PIE period, including the PIE cognates of wheat and barley, despite there being no evidence of agriculture in their homeland on the steppe. Moreover they managed to impose this vocabulary on the advanced Bronze age civilizations of Central and South Asia that had been practicing agriculture including the cultivation of barley and wheat for several millenia.

This is a most absurd proposition and it massively invalidates the steppe as the point of origin of the Indo-Europeans and the Indo-Iranians. The lack of agriculture on the steppe may ultimately lead to the abandonment of the Pontic-Caspian steppe as the PIE or early IE homeland.

 

Indo-Iranians and Agriculture

We may note that rather remarkably, for a group that was traditionally condemned to have been pastoralists in comparison to their ‘agricultural’ European cousins within IE, not only have the Indo-Iranians substantially preserved IE terms of agriculture, they have also managed to preserve the PIE terms for wheat and barley which are absent among the European branches.

Wheat

-> For wheat, we have in Sanskrit ‘godhuma’ and in Avestan ‘gantuma‘, Persian ‘gandum‘ which are cognate terms with Tocharian B ‘kanti‘ (bread) and Hittite ‘kant’, Luwian ‘kanta‘ both meaning wheat. This word being present in two of the earliest separated groups such as the Hittites and Tocharian, indicates its great antiquity dating to the PIE period while its presence in Indo-Iranian and Tocharian also ensures that it is unlikely to be a loanword in Hittite from a Near Eastern substrate. The cognates of this word are, however, not present in any of the European IE languages.

The argument, made by Mallory & Adams (2006) and Mallory (2014), that it must be a loanword from a non-IE Asiatic language source rings quite hollow. The Anatolians (Hittites, Luwians & Palaics) and Tocharians (A & B) separated quite early from the Indo-Iranians and there is zero evidence that the Anatolians ever had any later geographical proximity with the Indo-Iranians of Central and South Asia. Therefore, it is highly unlikely that Indo-Iranians and Tocharians in South and Central Asia would somehow use the same loanword for wheat which the Hittites and Luwians used in Anatolia several thousand miles to their west. Which culture and language in Asia could it possibly be that could loan this word for wheat to such widely separate groups ?

If this indeed was a loanword, the Hittites and Luwians were likely to have received it from the Hurrians and Hattic speakers of pre-Hittite Anatolia who were already cultivating wheat. Likewise the Indo-Iranians would naturally have borrowed it, if it was a loanword, from either the BMAC or the Harappan languages. Why would they borrow it from an altogether different source which moreover loaned the word to all the 3 geographically separate groups Therefore, there are solid grounds to believe that this cognate word for wheat in Indo-Iranian, Anatolian and Tocharian dates from the PIE period.

-> Mallory & Adams (2006) point to Hittite ‘sep(p)it’ as another likely PIE term for wheat but which is absent in other IE languages. However, a quick search on the online Sanskrit English dictionary, reveals that there is indeed a word ‘supistam‘ in Sanskrit which means wheat flour, which is most likely cognate with the Hittite term. It is quite likely therefore that Sanskrit has also managed to preserve this very archaic PIE term along with Hittite.

-> Gamkrelidze and Ivanov (1995) point to another possible word for wheat shared between Greek, Sanskrit and Balto-Slavic – Greek ‘puros’, Sanskrit ‘pura‘, Lithuanian ‘purai‘. However, its absence in Hittite, Tocharian and the other European branches makes it unlikely to be as old as the PIE period but it is still significant in the sense that the Indo-Iranians and some of the European branches seem to share a late IE word for wheat.

Barley

Coming to Barley, we have another cognate which is shared between the Hittites and the Indo-Iranians – Hittite ‘ewan‘ – barley, Persian ‘jav‘, Sanskrit ‘yava‘, Greek ‘zeiai‘ (wheat) and Lithuanian ‘jawai‘ (grain). It is derived from PIE verbal root *yeu ‘ripen, mature’. This word is again absent from most of the European IE languages.

There is another cognate which has in Sanskrit ‘sasyam’ (grain,fruit), Avestan ‘hahya’ (providing grain), Hittite ‘sesa(na)’ (fruit), northern Welsh (Celtic) ‘haidd‘ (barley) and Ligurian (Italic) ‘asia‘ (rye). This is another likely PIE word as can be seen from its distribution. Here again, besides the cognate term, the Indo-Iranians have also managed to preserve the same meaning of the cognate as in Hittite while others have transferred the meaning to something else.

Millets

In contrast to the words for barley and wheat, there is a word for millet which is shared between Germanic, Italic and Indic.  Northern High German ‘hirse‘ (millet), Latin ‘ceres’ (bread, grain), Kalash ‘karasha‘ (millet) and perhaps Sanskrit ‘kharA’/’kharAgari’ (type of millet plant). It may be noted that recent evidence indicates that the arrival followed by extensive cultivation of millets in Europe began only in the 2nd millennium BCE, likely from Central Asia. This may explain why the term for millet managed to survive in European languages unlike the cognates for wheat and barley which were already cultivated in Europe for long by the pre-IE people of that continent.

‘Millet’ as either an original meaning or a specific designation of a more generic word for ‘grain’ is interesting since it is not normally assigned to the early Neolithic package that entered Europe from the Near East but may have rather originated in central or east Asia (it is also found in the Harppan culture of India) and entered Europe across the steppelands. (Mallory & Adams, 2006).

Still another term is Sanskrit ‘dhAnya‘ (grain, cereal), Lithuanian ‘duona‘ (bread), Persian ‘dana‘ and Tocharian B ‘tano‘ (grain). This may or may not be PIE.

So quite extraordinarily, the Indo-Iranians have managed to preserve several original PIE terms for barley, wheat and other grains many of which are otherwise preserved only in Hittite, the earliest separated IE language. On the contrary, it is the European IE languages who have not managed to preserve the PIE cognate terms for most of the cereals, likely due to the influence of the pre-IE agriculturalists of Europe, the exception being millets which were only introduced in Europe in the 16th century BCE as per the latest evidence. This is an extremely important point of reference in locating the Indo-Iranian homeland.

Agriculture among the Harappans

In contrast to the steppe, agriculture among the Harappans was quite complex and advanced and they cultivated a wide variety of crops including wheat, barley and millets.

The Harappans grew lentils and other pulses (peas, chickpeas, green gram, black gram). Their main staples were wheat and barley, which were presumably made into bread and perhaps also cooked with water as a gruel or porridge. In some places, particularly Gujarat, they also cultivated some native millets; possibly broomcorn millet, which may have been introduced from southern Central Asia; and by 2000 BC, if not before, African millets. They fed local wild rice to their animals and probably began to cultivate it, though rice does not become an important crop until Post-Harappan times. The Harappans must have eaten a range of fruit, vegetables and spices : these included a variety of brassica, brown mustard greens, coriander, dates, jujube, walnuts, grapes, figs; many others, such as mango, okra, caper, sugarcane, garlic, turmeric, ginger, cumin and cinnamon, were locally available and probably grown or gathered by the Harappans, but the evidence is lacking. Sesame was grown for oil, and linseed oil may also have been used. (Harappa.com)

 

Conclusion

How did a group such as the Indo-Iranians, who supposedly originated on the steppe where there is no evidence of cultivation of either of the major crops or even of rudimentary agriculture, manage to achieve the feat of preserving the Proto-Indo-European terms of agriculture including that for wheat and baley, after having overrun the supposed non-IE speaking advanced urban agriculturalists of BMAC and Harappan civilizations ? How does any of this even make sense ?

The data would only make sense if either the Indo-Iranians had moved the least farther away from their PIE agricultural homeland or if in the case of their migration, they had moved into a region that did not quite have an agriculture as advanced as their PIE homeland.

Since, the agriculture practiced by the Harappans was quite advanced with multi-cropping as well as cultivation in both summer and winter, it is unlikely that there was any civilization around, much less on the steppe, that could have managed to completely or even partially impose its agricultural vocabulary on them. It is therefore more appropriate to suggest that the Indo-Iranians were those Indo-Europeans who moved the least distance away from the agricultural PIE homeland, in contrast to their European cousins. This implies that the PIE homeland must have been in a place close to the present inhabitation of the Indo-Iranians.

A similar argument in case of the Anatolian branches cannot be sustained since we have unequivocal evidence of substantial substratal influence on the Anatolian languages of IE from non-IE Hattic and Hurrian languages while no such pre-IE language has ever been detected in the traditional Indo-Iranian homelands of NW India, Eastern Iran and South Central Asia.