Climatic Factors in the Gupta Decline & After

“̍..[a] history whose passage is almost imperceptible, that of a man in his relationship to the environment, a history in which all change is slow, a history of constant repetition, ever recurring cycles..”

~ Braudel, as quoted in Jha, 2014: 588

Decline of an empire is never a linear process and involves complex to and fro of multiple factors until it eventually reaches a point of no return. The usual explanation for decline of the mighty Gupta Empire is explained away with two-pronged reasons – external invasions i.e. the Hūṇa attacks and internal deficiencies i.e. weaker monarchs after the illustrious Skandagupta. Some scholars go ahead to include weakening of the economy which is usually explained as a result of decline in external trade with the Roman Empire, though this has been a point of debate for some time. However, a significant factor typically conspicuous by its absence is climate.

Table of Contents & Links

Introduction
The Paradox
Palaeoclimatology
Correlation with the Dark Ages Cold Period
Floods in the Eastern Regions of the Gupta Empire
The Nile Flood Data
Climatic Factor in the Hunnic Movements
Volcanic Eruptions
Cosmic Body: A Possibility
Indian Literary Sources
Conclusion
Notes
Bibliography

§ Introduction

Relevance of environmental factors vis-à-vis socio-political history is a relatively new approach. Although with developments in the scientific world, its importance has been steadily increasing, it is still a rather underexplored facet of Indian history. Some esteemed scholars have given it its due importance¹ and there has been significant research in palaeoclimatology as well, but in my opinion it is not given its due attention. In this post, I will try to collate opinions, views and researches of various scholars pertinent to this issue and will try to understand the role of climatic factors regarding the decline of the mighty Gupta-s and will also tackle the question of why successor dynasties of the Gupta-s transitioned into monarchies whose structure was evidently different than that of the Gupta-s or the Maurya-s. They created kingdoms and empires but the pattern had changed radically with feudal overtones.

A significant change in the system of administration can also be seen – the start of which had been made during the Gupta-s but it became mainstay during the successor states. These successor states spanned their rule over large territories but they either could not match up the expanse of the Gupta Empire or even if they did attempt to reach that extent, they could only hold it for much lesser time relative to the Gupta-s. Did environment play any role in their expanse and pattern of administration? Something had definitely changed.

Were Gupta monarchs, particularly of post-Skandagupta period and the rulers that came after them fighting a lost battle against a force much larger than them? And to what extent did it affect their very existence? Why monarchies in the post-Gupta period unable to recreate the might of their predecessors. Let us try and find out.

§ The Paradox

In order to understand and analyze the role of climate in the decline of the Gupta dynasty and after, we first need to grasp what this decline entails and the paradox it encompasses in Indian archaeology. When an empire flourishes, its people also usually prosper. And this is exactly what the Gupta Empire generated – prosperity. India in the Gupta times saw immense development in the fields of religion, arts and culture, literature, scientific inquiries, economy etc. – truly a Golden Age and yet as I said, it shows a paradox. Archaeologically, it is considered poorer than the preceding Kuṣāṇa period. Generally accepted stance explains it as the decline of urbanism – a phenomenon which is said to start in early 4th century and intensifying towards the end of Early Historic Period (c. 500 BCE – 500 CE). It is then correlated with the rise of feudalism, suggesting a decline in centralization of the Maurya Empire kind. The theory was propounded by historians like R. S. Sharma in 1972 on the decay of Gangetic towns (Kennet, 2013: 331-332) and by B. D. Chattopadhyaya in 1974 (Kennet, 2013: 332). Since then, a two-stage de-urbanisation of India — first, after the 3rd century and the second, after the 6th century CE (Kennet, 2013: 332) — has been a standard in both archaeology and history of the Indian sub-continent.

But what explains this paradox, and does a paradox even exist? This will be tackled first before we begin to understand environmental factors of the period. Generally, excavations of have shown that many significant archaeological sites that had prospered till the Kuṣāṇa period show deterioration in the Gupta period and wherever habitation continues, it does not reflect the prosperity of the earlier period (Dhavalikar, 1999: 2). The number of sites occupied up to the 3rd century CE reduces to almost half by the time of the Gupta-s (Dhavalikar, 1999: 3). Eminent Archaeologist M. K. Dhavalikar had noted that deterioration in the quality of life is also seen in the Gupta period habitations and this seems almost a pan-India phenomenon (Dhavalikar, 1999: 3).

Important to note that according to Thakur, some sites in the peripheral areas like the lower Gangā valley, Oḍishā, Assam, Madhya Pradesh and Gujarāt witnessed strengthening of the urban tradition as oppose to decline in the core area of the Gangā valley in the Gupta period (Thakur, 1997: 15, 18). There were some sites like Mathurā (Dhavalikar, 1999: 3) and Kannauj (Kānyakubja) that showed continued habitation in the Gupta period but had shrunk in comparison to its previous Kuṣāṇa period. This curious case of Kannauj especially contradicted the description of a prosperous city which Hieun Tsang had mentioned about Harṣa’s capital when he had visited in the 7^th century (Dhavalikar, 1999: 4).

According to Dhavalikar, it is possible that during the Gupta and Harṣa times, habitation was located in some other parts of the ancient site but he also points out that it had nevertheless shrunk in the post-Kuṣāṇa period (Dhavalikar, 1999: 4).

According to the Chinese traveler, even Vaiśālī by the 7th century had been occupied by only a few inhabitants, the lakes had dried up and forests had been destroyed (Dhavalikar, 1999: 4). Ahicchhatra in Uttar Pradesh and many sites in Rājasthān did not fare better either despite the fact that numerous Gupta period terracottas had been found from these places (Dhavalikar, 1999: 5). Kapilavastu, Śrāvastī, Kuśīnagara, Champā (Bhāgalpur) and Charitravan (Chirānd) (Dhavalikar, 1999: 5) — all showed decline as per archaeology (Dhavalikar, 1999: 3-4). Except Nālandā, Vikramaśilā and some other sites in Bengāl that showed continuous habitation up to Pāla times (Dhavalikar, 1999: 5), deterioration was seen in other eastern Indian sites as well. Similar is the case in sites of Madhya Pradesh barring one – Ujjain (Dhavalikar, 1999: 5). And oddly enough, even South India presents a more or less similar picture (Dhavalikar, 1999: 5-6). It is possible that as many current villages are situated on late medieval habitations (Dhavalikar, 1999: 3), this could be a reason in confusing our understanding but there seems to be more to it.

The kind of archaeological evidence from the Gupta period that we have and magnificence of the Gupta Age that we know about cannot exist together. Puzzled by this contradiction, Dhavalikar questioned why habitation stopped at the Kuṣāṇa period particularly? And if it was to stop there, then why was it prolonged to 300 CE; why not stop at 250 CE? In his opinion, the entire Gupta Age needs to be considered in two separate phases – the earlier period i.e. 260-450 CE and the later Gupta phase i.e. 450-600 CE and it’s in the latter phase when relative decline sets in (Dhavalikar, 1999: 6). He explains that what we call the Kuṣāṇa levels in archaeology includes the early Gupta period as well and thus prosperity of these sites could be easily extended up to the 450 CE (Dhavalikar, 1999: 6), thus covering the Golden Age of the Gupta-s. Almost every evidence used to date these sites to the Kuṣāṇa period — the Kuṣāṇa coins, the pottery (Red Polished Ware) and the terracottas, even the size of the bricks — all continued to be used in the Gupta period particularly in the early phase, and many a times, even later (Dhavalikar, 1999: 7).

Derek Kennet is of the opinion that there’s also an enormous potential for selectivity and error in the reports of many of these excavated sites (Kennet, 2013). He mentions that many of these reports appear only in the form of short summary in the journal of Archaeological Survey of India — ‘Indian Archaeology: A Review’ (Kennet, 2013: 334). Another issue is that in Indian archaeology, conducting deep stratigraphic excavations had been a norm ever since the time of Sir Mortimer Wheeler, the main aim of which was and has been to ‘open up’ the chronology (Kennet, 2013: 334). Thus, the practice of taking the entire horizontal extent of occupation across the whole site into consideration did not really take root. As Kennet explains, these excavations then do not necessarily give a representative view of the whole of the area’s horizontal development, especially if there is a pattern of shifting centres which is usual in case of such extensive sites with multiple occupation mounds (Kennet, 2013: 334). This point was mentioned by Dhavalikar as well – the aim of especially earlier excavations was — ‘to reach the proto-historic Harappā levels for building up cultural sequences,’ he said (Dhavalikar, 1999: 8). This resulted in a hasty lumping together of historical levels of Śunga-Kuṣāṇa-Gupta period. Many sites were also put into the category of the Maurya-Śunga-Kuṣāṇa and therefore in the north, Gupta levels were not properly identified (Dhavalikar, 1999: 8).

In fact, as Dhavalikar noticed, the material aspects from the Kuṣāṇa period transition into finer quality in the early Gupta phase. All seems to suggest that decline actually set in later Gupta and post-Gupta phase. The Hūṇa invasions were significantly responsible for a lot of this wanton destruction. They decimated many buildings – religious and secular – all and that is why many prosperous towns that had continued their existence from Kuṣāṇa periods into the early Gupta phase discontinued habitation in the later phase of the Gupta Empire. And when habitation began again, it was marked by economic deterioration, suggested by construction of new buildings with Kuṣāṇa (or early Gupta period?) bricks (Dhavalikar, 1999: 7).

“This happened everywhere in the Indo-Gangetic plains and in Central India which was the core Gupta territory. Beyond this in the lower Gangā basin, the Gupta and post-Gupta habitations were in a somewhat better state (Dhavalikar, 1999: 7).”

Another point that needs attention is urban archaeology. Kennet states that unlike parts of Europe after the late 1940s, where “urban archaeology” was taken up fairly well, it is still largely unknown across the subcontinent for various reasons. As many of the important sites which can help us recreate a proper chronology continue to be occupied to the present day, the Early Medieval and later occupation history of many sites remain completely obscured from archaeological investigation (Kennet, 2013: 336). In dating of many of these sites, coins had been a major source, sometimes even the only source. And even this is not without its problems. The Kuṣāṇa coins being used in even in later periods has been mentioned above. Kennet mentions on the authority of Shastri that in the Vākāṭaka ruled territory, coins of earlier dynasties continued to be used, some of which were hundreds of years old (Kennet, 2013: 345). In another example, it is also possible that levels from many excavated sites that are published as Sātavāhana because they have yielded numerous Sātavāhana coins are in fact result of Gupta or post-Gupta occupation (Kennet, 2013: 345-346). This presents a whole new kind of possibility with regard to the later Early Historic and perhaps even Early Medieval periods in India.

“It is almost certain that coin residuality has caused layers at many sites in the region to be wrongly dated in exactly this way. Residuality and the relatively low numbers of post 4th century coins in circulation renders layers of the 4th century and later effectively invisible to the archaeologist, unless he or she is fortunate enough to have stumbled across one of the rare, datable coins or other objects of the Early Medieval period (Kennet, 2013: 345).”

Even considering these issues, there is however no doubt that in the later-Gupta or later Early Historic period a relatively low number of coins seem to have been in circulation and this relative paucity continued into the Early Medieval periods over large parts of the subcontinent. Kennet mentions low-denomination bronze coinage in this regard that is most commonly found on excavations (Kennet, 2013: 342) suggesting a dramatic decline in the deposition of low-denomination coinage in or after the 4th-5th century. Not only this but the gold coins of Skandagupta and the later mainline Gupta rulers (not to be confused with the Later Gupta-s) show a significant reduction in their gold content (Dhavalikar, 1999: 6). Thakur speaks on the gradual weakening of the long distance trade from 3rd century CE and agrees that post-Kuṣāṇa decline in trade with Central Asia did affect the urban prosperity and so did the reduction in trade with the Roman Empire, first in the 3^rd and then in the 6^th century CE (Thakur, 1992: 97). Dhavalikar and other scholars also mention the importance of external trade with the Roman Empire that had reduced. The external trade was affected and internal trade suffered as well which is suggested by the debasement of coinage in the later Gupta phase and by sheer paucity of coinage in the post-Gupta period (Dhavalikar, 1999: 8). Even the increasing use of cowries as the currency in the post-Gupta phase suggests the economic deterioration of the period.

There is no doubt that this trade had relatively declined but some scholars like Thakur, Kennet etc. question the level of importance that is given to the external, particularly Roman trade and its role in affecting the Indian economy extensively. Thakur points out that if this was the sole determinant of urban growth then how these towns emerged to prominence in the 7^th-6^th century BCE (Thakur, 1992: 98). And he is of the opinion that this also goes against the internal dynamism of the Indian system. This also would not be the first time when Roman trade in India was on the decline. Similar situation had arose with the Antonine Plague of the mid-to-late second century. Kennet is also of the opinion that there has been too much emphasis on Roman trade and its economic significance to the exclusion of trade with other regions. The trade with Western Asiatic countries like Iraq, Iran and Arabia was at least as significant, if not more to the Early Historic and Early Medieval economy of India. This is proved by the high proportions of Western Asiatic pottery on sites along the west coast and inland India as well as large quantities of Indian pottery on sites of this date in the Gulf (Kennet, 2013: 349).

Analysis of the local trade networks in the Early Historic Period in the subcontinent shows that the over-all impact of long distance exchange on its socio-political constructs may be over estimated, especially in case of exotic commodities.
(Smith, 2002)

However, this does not mean that it was a smooth road in the later-Gupta and the Early Medieval period. As stated before, this period seems to signal transition at the very least and decline at the worst. What it definitely shows is a period of instability. Kennet is right in his assessment when he concludes that the urban decline may not have been as abrupt or as absolute as believed, but the evidence suggests that this was nonetheless, a difficult and unstable period. He defines it by three main features – first, coin deposition in occupied areas was dramatically reduced that lead excavators to date late Early Historic layers too early. Second, many sites declined in size that lead to many excavation trenches being located in the abandoned areas once occupied by larger Early Historic sites. And third, the centres of many towns shifted that lead excavators to fail to locate medieval occupation buried under modern settlement (Kennet, 2013: 348). All these points mentioned point to some significant and fundamental structural changes. So what was the cause of this instability, that too for such a long period? If external trade could not have played a major role in it, then what explains the phenomenon? This is where analyzing climatic factor becomes even more important.

§ Palaeoclimatology

Before proceeding further, a basic understanding of some important terms and processes is pertinent which will help us understand the climatic aspect in a proper light. I have used Wikipedia here in this section for some basic definitions:

The main field of study relevant to this is palaeoclimatology which is the study of climates for which direct measurements were not taken. As instrumental records only span a tiny part of Earth’s history, the reconstruction of ancient climate is important to understand natural variation and the evolution of the current climate. Paleoclimatology uses a variety of proxy methods from Earth and life sciences to obtain data previously preserved within rocks, sediments, boreholes, ice sheets, tree rings, corals, shells, and microfossils. Thus, climate proxies are preserved physical characteristics of the past that stand in for direct meteorological measurements and enable scientists to reconstruct the climatic conditions over a longer fraction of the Earth’s history.

“The relationship of these proxies with climatic parameters is cardinal for the reconstruction of past climate and often contemporary data are used to calibrate the relationship. Further, their response time and thresholds to climatic shifts can range from being near instantaneous to periods spanning few centuries to several millennia (Singhvi & Kale, 2010: 3).”

Special climatic enquiries with regard to India also involve studying phenomenon of the Indian Summer Monsoon (ISM). It is an integral part of the Asian Summer Monsoon (ASM) system and is an annual climate manifestation of the seasonal reversal of cross-equatorial winds, which result in widespread precipitation over the Indian subcontinent during the months of June–September. It provides the major share of total annual rainfall of the country and has enormous impacts on the Indian economy. Paleoclimatologists have also studied it using climate proxies and their researches provides very important data that helps to understand climate variations in the subcontinent and their correlation with major epochs in Indian history. How important ISM is can be understood by this brief excerpt :

“Monsoon is the lifeline of the agrarian economy of the country. Though it is a stable atmospheric system and has never totally failed, erratic distribution of precipitation and, its spatial and temporal distribution can contribute to significant changes in food production and hydrological systems. Thus, for example, a reduction/enhancement in total amount of rainfall by a mere 20 to 30% in a season can cause agricultural droughts or severe floods. Even minor shifts in rainfall pattern (± 10%) during the monsoon season can significantly affect the agricultural productivity and water availability. The summer crops directly depend on summer monsoon rainfall and the winter crops in a major part of the country also depend on the residual moisture left by the summer rainfall. The irrigation system also depends on the water accumulated from the summer rains (Singhvi & Kale, 2010: 10).”

§ Correlation with the Dark Ages Cold Period

Studies have shown for long that the period from 400 to 1000 CE was characterized by adverse climatic conditions in the whole world (Dhavalikar, 1999: 8) and they obviously impacted Indian climate as well. Dhavalikar was one of the earliest proponents of analyzing environmental factors in post-Gupta decline and after. He mentioned that whenever temperate lands enjoy a warmer climate, tropical countries like India receive a good rainfall and when temperate lands cool, tropical countries suffer droughts. And that is why when Europe enjoyed a warmer climate from around 5th century BCE to 4th century CE, India also enjoyed a good rainfall during this period as confirmed by Arthaśāstra (Dhavalikar, 1999: 10). But with time, more detailed studies have been conducted that show correlation between global climatic periods and with Indian climatic changes across time. Important to note that climatic factor and its effects were even more pronounced in the past as societies and economies were more vulnerable to environmental changes.

A research was recently conducted to study the Holocene Indian Summer Monsoon (ISM) variability for understanding the spatio-temporal dynamics of the ISM precipitation (Quamar, 2022). This study presented a Late Holocene pollen proxy record of the changes in the ISM intensity from a 1.8 m deep lacustrine sedimentary deposit in Jammu District (Jammu and Kashmir State) in India. The results showed that ISM rainfall intensity was linked with some of the global climatic trends. Between ~3205 and 2485 cal yr BP [1255 – 535 BCE], mixed broad-leaved i.e. conifer forests occurred in the region under a warm and humid climate, probably indicating high monsoon precipitation. Subsequently, the conifers, such as Pinus sp., Cedrus sp., Abies sp., Picea sp. and Larix sp. increased comparatively and show dominance over the existing broad-leaved taxa between ~2485 and 1585 cal yr BP [535 BCE – 365 CE] under a cool and dry climate with reduced monsoon precipitation. Quamar found that the climate further deteriorated (towards attaining aridity under reduced monsoon condition) during ~1585 to 865 cal yr BP [365 CE – 1085 CE], which coincides with the Dark Ages Cold Period (DACP: CE 400–765; 1185–1550 cal yr BP) (Quamar, 2022: 47). Since ~865 cal yr BP (CE/AD 1085 onwards) to Present, the broad-leaved taxa, such as Alnus sp., Betula sp., Ulmus sp., Carpinus sp., Corylus sp. and Quercus sp. started expanding and showed their dominance over the conifers, indicating a warm and humid climate in the region with increased monsoon precipitation. This phase partly corresponds with the Medieval Climatic Anomaly (MCA) between CE/AD 950 and 1300. Important to our period under analysis, Quamar’s study found that mixed conifer/broad-leaved forests flourished in the region under cool and dry climate with reduced monsoon precipitation in this phase, which can be correlated with the Dark Ages Cold Period (DACP: CE 400–765: 1185 and 1550 cal yr BP) (Quamar, 2022: 44-45). Human activity and the pace of cereal-based agricultural activity also decreased comparatively during this time period (Quamar, 2022: 45).

Thus ISM precipitation regime showed correlation with the global climatic trends suggesting that Gupta and especially post-Gupta period up to 11^th century was a period of reduced monsoon.

Another study showed that significant shifts in monsoon rainfall seem to have occurred in concert with changes in the Northern Hemisphere temperatures and the discharges of the Himalayan rivers. The close temporal relationship between these large-scale hydro climatic changes and the intervals marking the significant sociopolitical developments suggests a plausible role of climate change in shaping the important chapters of the history of human civilization in the Indian subcontinent (Kathayat et al. 2017). It mentioned results from previous study of climate proxies created from the Sahiya Cave in Uttarākhand, India. This study also found decreased ISM strengths during the Dark Ages Cold Period (Kathayat et al. 2017) i.e. our period of study.

Another interesting research done some two decades back involved the study of ancient texts such as Arthaśāstra, Nirukta, Rāmāyaṇa etc along with modern scientific data which showed that periods of water stress as inferred from these texts seem to correlate with periods of low monsoon winds as inferred from oceanic records (Pandey et al. 2003; Singhvi & Kale, 2010: 14) This study by Pandey et al. created proxies for the monsoon winds for the past 4500 years using per cent fossil Globigerina bulloides in the Arabian Sea. The study found that weak summer monsoon winds are indicated by low abundances of the plankton Globigerina bulloides in the Arabian Sea. It was their hypothesis that when aridity increased in the region as is evident from palaeoclimatic studies people intensified rainwater harvesting as is seen from archaeological and historical evidences. The study found its confirmation as the period of severe monsoon failure coincided with details of construction of tanks and ponds mentioned by Varāhamihira in his Bṛhat Saṁhitā (Pandey et al. 2003: 48). Arid episodes correlated well with ancient Indian texts composed during those times containing descriptions of earthworks that may document aridity prevailing then (Pandey et al. 2003: 53). Therefore this study provides confirmation for the 6^th century as a period of severe aridity.

However, aridity is not the only concern for the period as there is also evidence for the other kind of natural calamity – floods. There is evidence of clustering of low-frequency, extreme floods between 400 and 1000 CE and into the most recent period (post-1950) (Singhvi & Kale, 2010: 18) which we will discuss further.

§ Floods in the Eastern Regions of the Gupta Empire

Recently, archaeologist Shanker Sharma of the Archaeological Survey of India presented his findings – “Challenging Stereotypes in Early Indian Buddhism” at an international conference organised by the Ludwig Maximilian University, Munich, in collaboration with the India International Centre (IIC) in Delhi in 2019. According to The Telegraph news reports (Raj, 2019), he found evidence for floods, particularly in the eastern regions of the Gupta Empire [Bihār and eastern Uttar Pradesh].

He found that many sites showed silt deposits between 0.6 and 2.5 meters thick, which suggests that these areas of the Gupta Empire were affected by large and disastrous floods. Balirājgarh (Madhubanī), Kolhuā, Rājā Vishāl Kā Garh and Chechar (Vaishālī), Titārā (Siwān), Chirānd (Saran), Panr (Samastipur), Kumhrār (Paṭnā), Champā (Bhāgalpur), Kushinagar and Siddhārthnagar (both in Uttar Pradesh) are some of such sites (Raj, 2019).

Sharma had been noticing abnormally high level of silt deposits corresponding to the time of the decline of the Gupta period while visiting various archaeological sites in Bihār and Uttar Pradesh, and focusing on it despite the fact that the sites are located in the middle of the Gangetic flood plains. Balirājgarh (Madhubanī) had a 7-metre-high fortification without any gap and people in the ancient times accessed it through a ramp. Sharma found that silt had crossed such high ramparts and around 1.7 metres of it was deposited inside. This suggested the severity of floods and gave an idea of how damaging this would have been to human settlements outside the fortification (Raj, 2019).

“The site of Kolhuā stūpa in Vaishālī district showed silt deposits of 1 metre to 1.5 metres, while the Chaumukhī Mahādev site in the same district had silt deposits of 2.5 metres. Panr site in Samastipur had a deposit of over 1 metre. There is evidence of high silt deposits at Kumhrār site in Paṭnā too. ‘There was no cultural occupation (proper civilisation) at the sites for several hundred years after. The populated areas were deserted. The growth achieved in the second urbanisation that had started in this region around 600 BC, centuries after the first urbanisation brought by the Harappan civilisation, was completely lost,’ Sharma said. The archaeologist also asserted that the deluge dealt ‘a death blow to the Gupta Empire’ that gathered its strength from Bihār region and ruled over a vast stretch from eastern India to northern and western India (Raj, 2019).”

Important to note that the Jain text Tilthogali Painniya [~ 575 CE] mentions about a king of Magadha called Caturmukha, a Buddhist who persecuted the Jain monks which resulted in a series of catastrophes in the city. The text says that it rained heavily and continuously for seventeen days, the water levels of the Ganges became dangerously high and there came a severe flood in the Sone, which engulfed Pāṭaliputra (Singh, 1975: 56). It also appears that there was a high toll of death. Kumharār excavations in 1912-1913 showed 820 feet deep silt and also a thick layer of ashes at the site of the 80 pillared halls (Singh, 1975: 57).²

Sharma is also of the opinion that these floods were also responsible in the decline of Buddhism. He said that “we need to find out how and why such deluge happened. Was there extreme rainfall or did any river get dammed due to landslides and then burst out to sweep the plains of Bihār and Uttar Pradesh? Did the rivers flowing in the area change their course due to an earthquake? These are the questions that need answers (Raj, 2019). ”

As of now, it is unsure at the moment what exactly was the cause of these floods and how are they connected with the data that palaeoclimatological studies have produced because if evidence from Sharma’s finds are to be considered along with the reduced monsoon suggested by studies mentioned above, then it seems that India at the time suffered both from floods in some regions and droughts in other. Whether they were happening simultaneously or in different time periods is also unsure. The evidence from literary sources may shed some more light on this as my limitation in knowledge of palaeoclimatology stops me from speculating any further. Whatever was the case, it sure seems to have been a period of climatic instability and a multi-disciplinary research on the period is necessary to make concrete assessments.

§ The Nile Flood Data

Before even more sophisticated researches were conducted to study the past climate in India, one important measure that was used to understand the past ISM variations was the Nile flood data (Dhavalikar, 1999: 12) which records the annual flood level of the river using a gauge Nilometer installed on the island of Roda in the Nile (Dhavalikar, 1999: 12). As for why the Nile floods are useful for studying past Indian climate? Dhavalikar explains:

“It must therefore be made clear that Egypt itself does not receive any rain, hardly one inch (2.5 cm) in the north; and in south Egypt which is a desert it normally does not rain at all. But the river is flooded every year as a consequence of the summer rainfall over the catchment basin of the Blue Nile in Ethiopia. It has been shown that north and east Africa get the same monsoon rain that we have in India. This was pointed out by Sir Gilbert Walker, a former Director General of Meteorology Department of Government of India, on the basis of his study of rainfall data from India and Africa from 1840 to 1910. He noticed a very close correspondence between Nile floods and Indian monsoon and has observed that – ‘It is now fully established that years of drought in western and north western India are almost invariable years of low Nile flood. The relation is further confirmed by the fact that years of heavier rain are also years of high Nile flood’ (Dhavalikar, 1999: 12).”

Important to note that north Africa and India experienced common intervals of droughts. A good example of it was the great Durgadevi famine of 1396 – 1408 CE and it is exactly during these years that the Nile levels were extremely low (Dhavalikar, 1999: 12). The data shows that the Nile had very low levels in late 7^th century and many more frequent intervals of low levels in subsequent 8^th-9^th century (Dhavalikar, 1999: 13) and this suggests the possibility of drought like situations during those periods in India as well. This is also corroborated by Indian literary sources which will be discussed shortly.

§ Climatic Factor in the Hunnic Movements

As mentioned previously, Hunnic invasions were also a major factor in the eventual decline of the Gupta-s. It is true that they were kept at bay for long by the Gupta-monarchs and a serious attack was foiled by Skandagupta when he gave them a resounding defeat in the year 457 CE. But it is also true that the danger of future possible Hūṇa attacks never went away and required much vigilance on part of Emperor Skandagupta on the huge western frontier of the Empire.³ The later Gupta monarchs were relatively weak resulting in another round of Hūṇa attacks under their leader Toramāṇa. Even though Toramāṇa was ultimately defeated by the Later Aulikara ruler Prakāśadharman, the destruction caused by him must have strained the Gupta-s further.⁴ This also stands true in case of Toramāṇa’s son Mihirakula, who was also eventually defeated firstly by the Gupta ruler Narsiṃhagupta Balāditya and then under the leadership of the Later Aulikara ruler Yaśodharman, but again it had made the Gupta positions weaker and also required great effort on part of the Indian rulers for a long time to keep the defeated Hūṇa-s from penetrating into the mainland India.

The point of mentioning the Hunnic invasions is that even the Hunnic movements, starting from their base in the Central Asian Steppes had changing climatic conditions as a major factor. The period even before the Dark Ages Cold Period [started ~ 5^th-6^th century CE] was seeing reduced precipitation and cool climate which increased its severity in the DACP further. Dendochronological studies [the science of dating events, environmental change, and archaeological artefacts by using the characteristic patterns of annual growth rings in timber and tree trunks] of the Dulan-Wulan tree-rings in Central China showed the existence of a severe drought of the fourth century that lasted nearly forty years, one of the worst in 2000 years (McCormick et el. 2012: 190).

Documented by the Dulan-Wulan tree-ring chronology, prevailing drought conditions began in 338 CE and continued until 377 CE, when wetter conditions returned (McCormick et el. 2012: 190). The study suggested that this drought played a critical role in driving the mobile pastoral federation that coalesced around the name of “Huns” somewhere east of the Don River, to seek pastures and predation farther to the west and south (McCormick et al. 2012: 190). The dendrodata thus confirmed speculation about an environmental factor in the Hunnic invasion that goes back at least a century. Historical sources indicate that the Huns reached the Don River by the 370s and crossed it c. 375 (McCormick et el. 2012: 190). A cycle of above average precipitation and then dry conditions in the 6^th century and the decades of drought therein could well have led to the westward migration of more pastoral peoples from Central Asia (McCormick et el. 2012: 199). This again provides strong evidence in understanding the climatic factor in the movement of Hunnic people that certainly affected the Gupta Empire, particularly in the post-Skandagupta period.

§ Volcanic Eruptions

Another significant climatic factor to consider is volcanic eruptions and their connection with climate cooling. Even though their effects on recent winters are modest in scale however, historically, they have been quite significant. But why is volcanic eruption important?

“When a volcano erupts, it spews sulfur particles called aerosols into the air, where they can persist for two to three years. These aerosols block out some of the sun’s incoming radiation, causing cooling. How much light gets blocked and how long the effect lasts depends on the location of the volcano and the magnitude of the eruption, as well as other variables in Earth’s natural climate-control system. Trees record the climate impacts of an eruption in the size of their rings—when a climate-related event occurs, the rings may appear wider or thinner than average, depending on whether the region is typically wet or dry and the normal length of the growing season. Meanwhile, the sulfur particles eventually fall to Earth and get incorporated into polar and glacial ice, providing a record of the eruptions (Zielinski, 2015).”

“Climate cooling is thus consistent with the idea that the atmosphere becomes shrouded in a haze of dust after large volcanic eruptions (Ball, 2008). Examples like the eruption of Tambora in Indonesia in 1815, of Mount Pinatubo in the Philippines in 1991 resulted in a dust veil, causing cooling and spectacular sunsets in the aftermath (Ball, 2008).”

How major role had volcanoes played can be gleamed from the fact that according to the data, nearly all extreme summer cooling events in the northern hemisphere in the past 2,500 years can be traced to volcanoes (Zielinski, 2015). Such volcanic eruption was considered the reason for the events starting from 536 CE. What happened in 536 CE? Byzantine historian Procopius of Caesarea wrote about the events of 536 – 537:

“And it was in that year that the greatest wonder of all took place: during the whole year the sun shone like the moon, without rays, as if it were losing its strength. It stopped shining clearly and brightly. From that moment war and plague and other calamities that brought death raged on and on among people” (Ganiev, 2014: 496).

Similar phenomena were noted in 536 – 537 CE in many other regions in Ireland, China, Chili, Europe and Asia (Ganiev, 2014: 496). The Greenland Ice Sheet Project (GISP) that studied samples of ice cores from Greenland and the Antarctic samples of the years 536, 538, 539, 541 and 543 from Greenland have a high content of sulfate, which speaks of its high content in the atmosphere in that period and of the low temperature of its formation. Dendrochronological studies from the Yamal Peninsula have also shown that the years 536, 537, 543 and 545 were the coldest with the most extremes (Ganiev, 2014, 497).

For some time it was considered that a volcanic eruption, probably from the Tierra Blanca Joven eruption, Ilopango in El Salvador was responsible for the cold decade centred around 540 CE in the Northern Hemisphere but latest research (Smith et al. 2020) has proved that the eruption could be pinpointed at 431 ± 2 CE in proxy records and shows that its impact was apparently limited. It appears to have had major effects on populations within ∼80 km of the volcano, where the regions were blanketed by decimeters of ash fallout and pyroclastic density currents (Smith et al. 2020). Other possible candidates have been the Krakatau in Indonesia or Tavurvur in Papua-New Guinea (Ganiev, 2014, 497) but the matter is not confirmed yet.

Actually, there is no widely agreed-upon single-source volcano eruption for the climate anomaly of the mid-6^th century and there is a possibility that the extreme cooling might have been due to multiple volcanic events. The resulting extreme cooling that was the phenomenon in the entire Northern Hemisphere therefore impacted India as well. Its indirect result on the sub-continent was its impact on the external trade with the Byzantine Roman Empire which suffered from devastating outbreak of the Justinian Plague — the pestilence even reaching the Sassanian Empire in the seventh century, severely affecting them (Sarris, 2022; Bonner, 2020: 194) and probably resulting in their ultimate defeat at the hands of the newly emerging religion of Islam.⁵ These volcanic eruptions seem to have severely affected China as well (Gao, 2021). The direct effects of the resulting climate cooling will be mentioned in a short while. The role that this plague played in the eventual rise of feudalism in Europe has also been noted by many scholars. It appears that results from climate cooling and the DACP were crucial for similar changes in India – thus producing fundamental changes in the socio-politcal system on a global scale. Interesting to note that India herself has not been unknown to effects of volcanic eruptions in the past — for example, the eruption of Mount Tambora in Indonesia in 1815 CE resulted in crop failures, famines and even pandemic of cholera and its transmission out of India.⁶^,7

§ Cosmic Body: A Possibility

There is another interesting possibility that some scientists have suggested. According to it, the 536 – 545 CE climate cooling could have been due to a cosmic body (Ganiev, 2014, 497). Dr. Dallas Abbott’s hypothesis is that the 536 – 545 climate cooling was caused by both volcanic and cosmic factors (Ganiev, 2014, 497). In support of this theory, he mentions that the specimens of Greenland ice have shown to contain particles of alien origin with a content of Ni-rich material and Fe oxide-rich spherules, which are characteristic for cosmic bodies. Some argue that this cosmic body might have been the Halley’s Comet, which appeared in 530 (Ganiev, 2014, 497). “In that year the comet appeared in its brightest form ever. The Byzantine chronicler Ioannes Malalas wrote:

“During the reign of Justinian I there appeared in the west a huge, terrifying star from which a white ray led upwards, causing lightning. Some called it a torch. It shone for twenty days, and there was a drought, and in the towns people were killed and many other terrible events took place” (Ganiev, 2014, 497).

At the moment, even though we cannot be sure of what exactly was the cause of this climate anomaly, it definitely created conditions conducive for the generation and swift expansion of the plague. The DACP that seems to have started in the early 5^th century had already caused crop failures and famines. This also seems to have made the expansion of the pestilence much faster. However, it seems that plague did not affect India much [contributing factor to this might have been our warmer climate and/or our better personal hygiene at the time relative to the rest of the ancient world]. But other effects from extreme cooling due to volcanic eruptions [or cosmic body?] in the mid-sixth century along with the general cooling of the time due to the Dark Age Cooling Periods resulted in reduced monsoon in some regions of the Indian sub-continent. On the other hand, we also learn that in some regions, there were instances of extreme floods between 400 and 1000 CE as mentioned above. Combined effect of these can be understood from some important events mentioned in the literary sources of this period which we are going to discuss next.

§ Indian Literary Sources

With time, data collected on this subject seems to provide confirmation that climate in late fifth and sixth century had worsened. Just as environmental changes were affecting the entire world, they were affecting India as well. Thakur especially mentions extensive floods and damage to the Sudarśana Lake Dam during the time of Skandagupta (Thakur, 1992: 100). Important to note that the Vāmana Purāṇa, compiled around 7th century CE in Kurukṣetra, informs that several holy places associated with big ponds felt helpless against the advancing dunes of sand (Thakur, 1992: 100).

Most important literary source for this period is the Bṛhat Saṁhitā of Varāhamihira [compiled ~ 6^th century CE] in which he makes 42 forecasts of famine, and 32 of drought and scarcity of rain; this in contrast to only 11 of heavy rainfall. The reduction in rainfall is also suggested by a sudden increase in the rituals aimed at pleasing the rain-god and innumerable references to famine in contemporary texts (Thakur, 1992: 101). The forecasts in the Bṛhat Saṁhitā textualise trends of urban decline in the core area of the Gangā valley and a strong urban continuity in the peripheral zones (Thakur, 1997: 15).

Thakur points out that while there are eleven prediction for the destruction of the middle country (madhya deśa) and fourteen to northern India, and its neighbouring zones, only one relate to the east, one to the Deccan, two to the South and five to the West. He is correct in pointing out that the vulnerability of the madhya deśa and northern India is too marked to need any comment.

There are numerous predictions to the destruction of the geographical zones of, Kuru, Pāñcāla, Kāśī, Kauśal, Magadha, Aṅga and Videha. The forecast about destruction of people living between the banks of Prayāga and city of Avanti also suggest ecological imbalance (Thakur, 1997: 19). Varāhamihira also predicts trouble for people living on the river banks in general and for those living on the banks of Indus, Yamunā, Venā, Sarayu, Sona, Chārudevi, Narmadā and Kāverī as well as those residing on the east coast in particular (Thakur, 1997: 21). The severity of the floods suggested by Shankar Sharma [mentioned above] thus finds some corroboration in this magnum opus of Varāhamihira.

Forecasts of drought in Bṛhat Saṁhitā are interesting in light of evidence from the Javanese Chronicles pointed out by Dhavalikar – a story in it states that a ruler of Gujarāt, who was warned in advance about the impending period of droughts, sent his son with 5,000 persons including cultivators, artisans, warriors, physicians etc. to Java in the 6th century CE (Dhavalikar, 1999: 10). Severe famines in 7th-8th centuries CE are also reflected in Daśakumāracaritaṃ [Ch. VI] of Daṇḍi in which he details a prolonged drought lasting for twelve years. This resulted in crops failures, medicinal herbs losing their curative power, trees boring no fruit and even cannibalism to assuage the hunger. Dhavalikar suggests the possibility that this same famine might have been mentioned in the Bṛhad Nāradīya Purāṇa (38, 87) of the same period which states that people will migrate to other areas because of famines (Dhavalikar, 1999: 10). Mārkaṇḍeya Purāṇa which also belongs to 6th-7th century refers to a hundred years’ drought during which people underwent untold sufferings because of prolonged famines (Dhavalikar, 1999: 10). The information from the Jain text about a devastating flood in Pāṭaliputra in late-6th century has been mentioned above. As noted before, the Nile flood data also shows low levels of the river during 687-696 AD. Dhavalikar believes that this might have been the period of the famine described by Daṇḍi.

“Extremely low levels of the Nile have also been recorded during 770-782 AD, 788-802 AD, 943-951 AD and 963-967 AD. Actually 770-802 AD appears to be one long phase of low Nile levels and we may not therefore be far off the mark if we infer that these were the years of droughts in India. This explains why there are descriptions of frequent famine in ancient literature of the post-Gupta period. The hundred year famine referred to in the Mārkaṇḍeya Purāṇa in its Devī Mahātmaya section may be the reflection of the adverse environment during this period (Dhavalikar, 1999: 13).”

It thus seem like northern India was the first to get affected by the changing environment in the late 5^th – early 6^th century and during this time, other parts in the sub-continent enjoyed relatively better conditions. However, this seems to have changed by 7^th century when climate disturbances were a pan-India phenomenon. In connexion to this, change in the pattern of paying the officials in this period is also significant. While in the Maurya period the officials were mostly paid in cash, in the time of the Gupta-s we see a transition towards giving land grants in perpetuity. And this system intensified by the late-Gupta period with permanent land grants to priests, temples and educational establishments (agrahara) along with official land grants; this finally resulted in land grants becoming the main system through which officials were paid by the time of Harṣavardhana of Kannauj as told by Hieun Tsang (Dhavalikar, 1999: 11).

Even though, it is noted that the role of declining Roman trade in the economy of this period is over-rated [at the expense of trade with Western Asia for example]; and studies have shown that internal trade in the subcontinent was still robust enough but archaeological finds do suggest a period of instability in late Early Historic – Early Medieval Period which is also characterized by paucity of coins in comparison to the previous period. This does suggest that the economy was suffering from instability and deterioration vis-à-vis the previous period. Added to it, the climatic factors resulted in crop failures and famines certainly would have reduced revenues of the Empire in the late Gupta period and those of its successor dynasties. We are not sure how a strong Gupta Emperor would have been affected by these disturbances but the cumulative effect of these during the vulnerable time later kings of the Gupta dynasty was a little too much to bear without any adverse effects. When we also consider that by this time, various vassals and/or hereditary governors of the Gupta-s had become strong enough to take advantage of weaker monarchs in late fifth – early sixth century, it is not that difficult to understand how adverse climatic factors must have acted as pressure points in their eventual decline. This also suggests according to Dhavalikar, that feudalism does not seem to have been a cause of the economic decline, rather it was a consequence of the adverse climatic factors (Dhavalikar, 1999: 11). And this aspect was certainly not peculiar to India as it also had consequences the world over. This is why land grants were increasingly made in the peripheral areas – vast stretches of land were available and hence land grants could be made (Dhavalikar, 1999: 11).

§ Conclusion

This study points towards the significance of analyzing the environmental factors and their agency in important events of history. It does not mean that they are monocausal in explaining either the rise or decline of any dynasty or other crucial events of the past. The human agency always remains just as crucial, if not more. The climatic factors might be more determining in some particular cases, but most of the times, as just mentioned, they act as significant pressure points, especially in case of prevailing social and/or economic instability which seems to have been the case in the eventual decline of the Gupta dynasty. It is also important to note that the climatic changes that the Dark Ages Cold Period initiated in the 5^th century, intensified in the 6^th century and the long term environmental stress therefore created extreme results in the entire world from which India certainly was not unaffected. The effects did not just stop at the Gupta decline but also proved salient in determining the structure of administration of the states or the monarchies in the post-Gupta period which then had significant permanent consequences on our social structure.

NOTES

⁽¹⁾The role of environmental factors in establishment of different kinds of political entities is an important matter. The role that it played in the rise of Magadha as the penultimate state in ancient India has been discussed by Jha in his Migration, Settlement, and State Formation in the Ganga Plain: A Historical Geographic Perspective (2014). ⁽²⁾Geographical situation of Pāṭaliputra played important role in its rise as well as its downfall. The city was uniquely encircled by four rivers (Singh, 1975). ⁽³⁾Read my post – Gupta-Hūṇa Relations: A Study for the analysis of encounters between the Gupta Empire and Hunnic tribes till the period of Emperor Skandagupta. ⁽⁴⁾For detailed description of Hūṇa ruler Toramāṇa in India and his final defeat by the later Aulikara ruler Prakāśadharman, see my post – The First Indo-Hunnic War. ⁽⁵⁾ In the year 542 when a Perso-Roman War was raging, Sasanian Emperor Khusro got infected and retreated to the highlands of Azarbaijan to recover his health. Bonner notes that if the sufferings of Ctesiphon resembled those of Constantinople, work of every kind would have ceased as the disease ran its course, and the Iranian capital may have lost nearly half its population. But it may be possible to connect some stray notices of the Persian royal tradition with the aftermath of a large mortality. For more information on this see (Bonner, 2020). ⁽⁶⁾Before Tambora, cholera was mostly confined to the Ganges Delta in India. But this eruption caused local climate changes, including the loss of one monsoon, that have been blamed for the emergence of a new strain of the disease. This strain was able to spread more easily, and between 1819 and 1820, it moved out of Bengal and into Burma and Thailand (then called Siam). By 1831, it had reached Western Europe, and it came to North America the following year (Zielinski, 2015. April 9).⁽⁷⁾The morning temperature of Madras in 1815 was 11 degrees Celsius on Monday, April 24, and by Friday, April 28, it had dipped to minus 3 degrees Celsius. There are unverified reports of snow falling too but that may be an exaggeration…The cause of this freak phenomenon was the volcanic eruption of Mount Tambora in distant Indonesia (see The Hindu report ‘When Madras Froze Over’, 2015).