In my previous article, I wrote about the first settlement of Kerala as described in Keralolpatthi, re-interpreted based on an Egyptian connection. Before I proceed to the next chapter, the reign of the perumal, I would like to digress a bit to discuss how I determined their timeline.
Ancient Egypt had several historical periods delineated by Egyptologists, among them: the Old Kingdom (ca. 2900 BCE -2180 BCE) when the Great Pyramids were built and the ports on the Red Sea saw frequent and numerous trans-oceanic expeditions; the Middle Kingdom (ca. 2055 BCE — 1650 BCE) when Egypt began collaborating with the Phoenicians on maritime trade; and the New Kingdom (ca. 1550 BCE — 1077 BCE), famous for many kings and queens in popular imagination, among them Akhenaten, Nefertiti, Tutankhamun, Hatshepsut and Ramesses.
Pharoah Ramesses II, who ruled during the New Kingdom, was embalmed with black pepper when he died around 1200 BCE, suggesting that black pepper, endemic to Kerala, was known to the Egyptians. Archaeobotanical evidence shows that agriculture began in Kerala towards the latter half of the third millennium BCE, suggesting that the land was populated by an agrarian society around this time. Thus, it is safe to assume that Kerala was settled sometime between 2500 BCE when agriculture began, and 1200 BCE when evidence of Kerala’s trade goods surfaced in Egypt.
According to Keralolpatthi, there were two different eras when the perumal ruled Kerala. These two eras were separated by a gap of several years during which time there was no colonial rule, and the settlers were left to fend for themselves. Incidentally, between 2500 BCE and 1200 BCE, ancient Egypt had two periods of extreme political instability and chaos: the First Intermediate Period (ca. 2180 BCE — 2055 BCE) between the Old Kingdom and the Middle Kingdom; and the Second Intermediate Period (ca. 1650 BCE — c. 1550 BCE) between the Middle Kingdom and the New Kingdom. It is quite likely that the period of instability in Kerala mentioned in Keralolpatthi coincided with one or the other of these two periods in Egypt.
Although Keralolpatthi provides no reliable absolute dates, it does specify with remarkable particularity the relative chronology down to the years and months for nineteen out of the twenty-five perumal. There were twenty-two perumal in the first era and three in the second era. Both eras were marked by certain distinct characteristics. First, direct centralized rule by the king through the perumal was not the default governing mechanism in either era: local chieftains had to specially request the king for central rule. This pattern resembled the decentralized governance of the Old Kingdom and early Middle Kingdom in Egypt more than the central rule of late Middle Kingdom or the New Kingdom. Second, while ambitious nation-building was the chief hallmark of the first perumal era in Kerala, characterized by laying down laws and building fortresses and temples, similar to the expansive projects undertaken during the Old Kingdom in Egypt, the second perumal era was geared for trade, and more tempered, as evidenced by joint venture with Phoenicians, evoking Middle Kingdom patterns.
A third characteristic was the turnover among the perumal: although each perumal was brought in to rule for a tenure of twelve years, during the first perumal era, only eight out of the twenty-two ruled for their full twelve-year term, whereas the second perumal era was stabler despite its overall briefness, as all three completed their full twelve-year term. This pattern perhaps suggests political turmoil in the motherland during the first perumal era and political stability during the second perumal era, indicative more of the turmoil of the late Old Kingdom followed by the stability of the early Middle Kingdom rather than the early stability and later turmoil of the New Kingdom.
Further, given the evidence of trans-oceanic voyages during the Old and Middle Kingdoms, and the dearth of it during the New Kingdom other than Hatshepsut’s expedition to Punt, it seems more likely that both the first and second perumal eras predated the New Kingdom. Therefore, the period of instability in Kerala must have been the First Intermediate Period, suggesting that the earliest settlers to Kerala arrived during the Old Kingdom, and the earliest perumal were sent by the Old Kingdom pharaohs.
The perumal in Kerala were appointed for a term of twelve years each, but many ruled less than their allocated time. A possible reason for the short reign could be a regime change in the homeland when a new monarch replaced his predecessor’s loyal officers with his own men. Thus, if Kerala and Egypt were connected politically with direct rule by the pharaoh, one of the ways such a connection would manifest is by synchronicity of regime change in both lands. In other words, when a new pharaoh ascended the throne in Egypt, a new perumal would be dispatched to the colony at around the same time.
A correlation between two different sequences of events, such as the accession of a pharaoh and appointment of a perumal, can be studied using a “spike train” analysis. A spike train is a list of times when certain measurable events have occurred. Spike train analysis is employed extensively in neuroscience to understand neural responses to stimuli within the human brain. Two neural spike trains observed within the same time window in the brain may be independently generated but may nevertheless correlate to each other because of underlying biological mechanisms. Spike train analysis helps to determine if there is correlation. This method can be applied to any spike trains, not just neural ones.
Various methods are available for determining correlation between spike trains, the most common one being computing a correlation coefficient called spike time tiling coefficient (STTC). STTC is a measure of the number of spikes in one train A that occurs synchronously with spikes of another train B. It is computed using the following formula:
PA is the proportion of spikes in A that lies within a time window of any spike from B, PB is the proportion of spikes in B that lies within the same time window of any spike from A, TA is the fraction of recording time T that lies within the time window of any spike in A, and TB is the fraction of recording time that lies within the same window of any spike in B. The STTC is a measure of “affinity” between the two spike trains A and B that speak to a common link of some sort. In general, a positive value for STTC indicates positive correlation (e.g., spikes certainly occur synchronously), a negative value indicates negative correlation (e.g., spikes certainly do not occur synchronously) and a zero signifies no correlation (e.g., spikes may or may not occur synchronously).
STTC can be useful in understanding correlation between the events in Egypt and those in Kerala to a certain extent. For example, regime change in Kerala can be represented as a spike train A consisting of the years of appointment of the perumal. Similarly, regime change in Egypt can be represented as spike train B consisting of the years of ascension of the pharaohs. A statistically significant value of STTC can, but not necessarily, indicate a correlation. One of the reasons that STTC alone cannot predict absolute correlation is artificial synchronicity necessarily introduced into the analysis because only the relative chronology of the perumal is known. Therefore, the years of their appointment must be tied by an appropriate assumption to the absolute chronology of pharaohs so that the two can be analyzed coherently. According to Keralolpatthi, the first perumal who was seated to rule for twelve years, ruled only for eight years and four months, after which he was recalled. If he was recalled because of a regime change, then the next perumal was appointed synchronously with ascension of the next pharaoh. The absolute year of appointment of the second perumal can now be pinned to the known year of ascension of this second pharaoh, and subsequent perumal tacked on according to the relative chronology in Keralolpatthi. However, this pinning adds artificial synchronicity to the two spike trains. From a mathematical standpoint, because the number of synchronous events is already non-zero, the STTC for otherwise uncorrelated spike trains is also non-zero and statistically significant.
One method to screen out this spurious STTC value is to compare the spike train representing perumal appointments against spike trains representing random variation of pharaonic accession events. With a large enough sample size of these randomly generated spike trains, the average STTC values settle to a baseline representing the single artificial correlation injected by pinning the second perumal to the second pharaoh. Any STTC value greater than this baseline signifies a positive correlation supporting the hypothesis that pharaonic accession events correlate to perumal appointment events.
A positive STTC value can also be substantiated by an appropriate political context that explains the correlation, given what is known about Egypt and Kerala. Thus, for example, Keralolpatthi states that the ninth perumal abdicated his seat voluntarily; therefore, this particular event is likely not correlated with a regime change event. In another example, it is likely that maritime expeditions were undertaken in Egypt during the Old Kingdom, but not during the First Intermediate Period; that such oceanic ventures continued during the Middle Kingdom but not later; that Egypt’s venture into open seas during the early part of the New Kingdom was intermittent, if at all, before Hatshepsut’s famous Punt expedition; that given Kerala’s black pepper found in the body of Pharaoh Ramesses II, the pharaoh who sent the first perumal must have ascended the throne much earlier; and given the cultural and political characteristics of the two perumal eras, it is more likely that the perumal were sent during the Old Kingdoms and Middle Kingdoms rather than during any other time in Egypt’s history. Hence, a pair of spike trains having a statistically high STTC that can be explained with such cultural, archeological, epigraphic, and other context surrounding the spike events may be reasonably found to be correlated.
Using an open-source numerical analysis software tool called “Octave,” I generated different spike trains corresponding to different pharaohs who were assumed to have sent the first perumal. I also constructed several alternate spike trains of perumal with different terms for those whose reigns were not specified in Keralolpatthi. Two-hundred random spike trains were generated and compared against the spike train representing the perumal to determine a baseline STTC for a singly correlated event. Any positive deviation from the baseline indicated a definite correlation. When I ran the analysis, not only was a positive correlation found, corroborating the hypothesis that the perumal in Kerala and the pharaohs in Egypt were connected, it was found from this STTC analysis that the first perumal was most likely sent to Kerala during the reign of pharaoh Sahure of the Old Kingdom, sometime between 2465 BCE and 2325 BCE.
This chronology as determined from mathematical analysis matched remarkably with the events and cultural milieu of Old Kingdom. For example, according to Keralolpatthi, the third perumal departed rather suddenly, claiming there was no one to protect his kingdom. This event coincided with the political turmoil around the accession of pharaoh Shepseskare, who may not have been the rightful heir to the throne of Egypt. Likewise, a conflict with the fourth perumal leading to his assassination may have reflected the political situation in Egypt with Neferefre’s short reign. The religious debacle suffered by the ninth perumal matched with the religious reformations during the reign of pharaoh Djedkare.
Between the twelfth and eighteenth perumal, six perumal circulated within thirty-six years, an average of half the expected tenure, reflecting the growing crisis of unitary monarchy in Egypt. The turnover of dignitaries in key administrative positions proceeded similarly at an astonishing pace under pharaohs Teti, Userkare, and Pepi I, whose reigns were marred by various scandals. During Pepi I’s reign, temples in far-flung colonies increased in significance to the Egyptian administration; reflecting this policy, the nineteenth perumal, sent around the same time to Kerala, built a temple there. Concurrently with the period of the Sixth Dynasty (ca. 2250 BCE) during which time many fortresses were built across Egypt, Kerala too experienced a flurry of fortress-building by the fifteenth perumal onward. By the time of the last perumal of this era, Egypt transitioned into the First Intermediate Period after the long reign of Pepy II.
During the First Intermediate Period, the Old Kingdom disintegrated into a number of small chiefdoms ruled by local governors, who were supposedly still under the king in Memphis but in reality, were mostly sovereign. This collapse of pharaonic authority put a temporary stop to Egypt’s colonial aspirations. There are very little archeological and textual records of the immediate period following the collapse of the Old Kingdom, but it is perhaps safe to assume that the political chaos in Egypt left no resources for any colonial enterprise at any scale. Pierre Tallet’s excavations at Ayn Soukhna and Wadi al-Jarf suggest that the operations at these ports that had started with Khufu in the Fourth Dynasty stopped during the First Intermediate Period, indicating that the massive trans-oceanic voyages of former times ground to a halt during this time. This period of uncertainty was echoed in Kerala, where the departure of the twenty-second perumal was followed by a long hiatus during which time the local chieftains were left to their own devices. It was a period of chaos and turmoil.
The end of this period, its duration unknown, was marked by a sudden revival of the perumal from the distant homeland. This reinstatement of pharaonic control coincides with the rise of the Middle Kingdom, for records indicate that Pharaoh Mentuhotep III revived the port of Mersa Gawasis on the Red Sea during the eighth year of his reign. Thereafter, it is likely that oceanic ventures undertaken by the Egyptians were in collaboration with Phoenicians, as attested by Keralolpatthi too. After around 1800 BCE, Egypt’s oceanic ventures stopped, as did the era of perumal in Kerala, suggesting that direct rule by the pharaoh stopped around this time. Control through a perumal was never again re-established according to Kerala’s local lore and epigraphic records. Kerala became a sovereign nation, its history diverging from Egypt’s thereafter, but its culture remained ossified in its early Middle Kingdom form even into the nineteenth century CE.
Having now fixed the timelines of the era of the perumal, in my next article I will go into the details of their reign as described in Keralolpatthi, tying the events in Kerala to the happenings in faraway Egypt.
Bibliography
Abd el-Raziq, Mahmoud, Georges Castel, Pierre Tallet, and Gregory Marourad. 2012. “The Pharaonic Site of Ayn Soukhna in the Gulf of Suez 2001–2009 Progress Report.” In The Red Sea in Pharaonic Times: Recent Discoveries along the Red Sea Coast. Edited by Pierre Tallet and El-Sayed Mahfouz, 3–20. Cairo: Institut Francais D’Archeologie Orientale.
Adams, William. 1984. “The First Colonial Empire: Egypt in Nubia, 3200–1200 B.C.” Comparative Studies in Society and History, vol. 26, no. 1: 36–71.
Bard, Kathryn, and Rodolfo Fattovich. 2011. “The Middle Kingdom Red Sea Harbor at Mersa/Wadi Gawasis,” Journal of the American Research Center in Egypt, vol. 47: 105–29.
Bárta, Miroslav. 2017. “Radjedef to the Eighth Dynasty.” UCLA Encyclopedia of Egyptology. https://escholarship.org/uc/item/67n4m4c4.
Bietak, Manfred. 2010. “From where came the Hyksos and where did they go?” In The Second Intermediate Period (Thirteenth–Seventeenth Dynasties): Current Research, Future Prospects. Edited by Marcel Marée, 139–82. Leuven: Uitgeverij Peeters.
Cutts, Catherine and Stephen Eglen. 2014. “Detecting Pairwise Correlations in Spike Trains: An Objective Comparison of Methods and Application to the Study of Retinal Waves.” The Journal of Neuroscience, vol. 34, no. 43: 14288–14303.
Dulíková, Veronika. 2016. “One of the minor gods: A case study on Khentytjenenet, an Old Kingdom deity of the Memphite necropolis.” Prague Egyptological Studies, vol. 17: 36–46.
Fuller, Dorian and Eleni Asouti. 2008. Trees and Woodlands of South India: Archaeological Perspectives. Walnut Creek: Left Coast Press.
García, Juan. 2013. “The Territorial Administration of the Kingdom in the 3rd Millennium.” In Ancient Egyptian Administration. Edited by Juan García, 85–154. Leiden: Brill.
Gee, John. 2004. “Overlooked Evidence for Sesostris III’s Foreign Policy.” Journal of the American Research Center in Egypt, vol. 41: 23–31.
Grajetzki, Wolfram. 2015. “Middle Kingdom History: An Overview.” In Ancient Egypt Transformed: The Middle Kingdom. Edited by Adela Oppenheim, Dorothea Arnold, Dieter Arnold, and Kei Yamamoto, 306–10. New York: The Metropolitan Museum of Art.
Hashemi, Alireza, Ashkan Golzar, Jackson E.T. Smith, and Erik P. Cook. 2018. “The Magnitude, But Not the Sign, of MT Single-Trial Spike-Time Correlations Predicts Motion Detection Performance.” The Journal of Neuroscience, vol. 38, no. 18: 4417–4399.
Keralolpatthi: The Origin of Kerala. 1868. [In Malayalam]. Mangalore: Pfleiderer & Riehm.
Lehner, Mark. 2013. “The Heit el-Ghurab Site Reveals a New Face: The Lost Port City of the Pyramids.” Aeragram, vol. 14, no. 1: 2–7.
Manzo, Andrea. 2011. “Punt in Egypt and Beyond: Comments on the Impact of Maritime Activities of the 12th Dynasty in the Red Sea on Egyptian Crafts with Some Historical and Ideological Thoughts.” Egypt and the Levant, vol. 21: 80–82.
Müller-Wollermann, Renate. 2014. “End of the Old Kingdom.” UCLA Encyclopedia of Egyptology. https://escholarship.org/uc/item/2ns3652b.
Strudwick, Nigel. 2005. Texts from the Pyramid Age. Atlanta: Society of Biblical Literature.
Tallet, Pierre and Gregory Marouard. 2016. “The Harbor Facilities of King Khufu on the Red Sea Shore: The Wadi al-Jarf/Tell Ras Budran System.” Journal of the American Research Center in Egypt, vol. 52: 135–77.
Tallet, Pierre. 2013. “Research Report: The Wadi el-Jarf Site: A Harbor of Khufu on the Red Sea.” Journal of Ancient Egyptian Interconnections, vol. 5, no. 1: 76–84.
Tallet, Pierre. n.d. “Archaeology — The Unique Testimony of the Cheops Papyri.” LaRecherche. Accessed on September 11, 2020. https://www.larecherche.fr/anglais/archaeology-unique-testimony-cheops-papyri-0.
Verner, Miroslav. 2002. Abusir: Realm of Osiris. Cairo: The American University in Cairo Press.
Ward, Cheryl. 2014. “Sailing the Red Sea: ships, infrastructure, seafarers and society.” In Ships, Saints and Sealore: Cultural Heritage and Ethnography of the Mediterranean and the Red Sea. Edited by Dionisius A. Agius, Timmy Gambin and Athena Trakadas, 115–23. Oxford: Archaeopress.
