The Residences of the Tarki Shamkhals: A Historical Tradition in Comparative Perspective

Abstract

This article addresses a live and contested question of Dagestani historiography: why the rulers of the largest late-medieval polity of the North-Eastern Caucasus — the Shamkhalate of Tarki — simultaneously maintained two residences. The problem is set against a broad comparative background of steppe, Islamic and European parallels.

Keywords: North-Eastern Caucasus; Shamkhalate of Tarki; residence; Kazi-Kumukh; Tarki (Targhu).

Two residences: the problem

It is common knowledge that almost every ruler has, alongside an official capital, particular residences of his own. Many of these names are familiar to the ear — it is enough to recall El Escorial and Versailles. In Dagestani historiography, however, the question of residences carries a particular resonance. It is still an open question which of the two residences of the rulers of the largest state of the late medieval North-Eastern Caucasus — the Shamkhalate — played the leading role: the summer residence at Kazi-Kumukh, or the winter residence at Tarki? Nor has the question yet been fully resolved of why the shamkhals chose precisely these two rather widely separated settlements as their seats. The literature on the problem is extensive. We shall not rehearse all the arguments on either side; instead, in order to address the problem, we shall give preference to the comparative-historical method. For all its particular features, the North-Eastern Caucasus of the 15th–17th centuries was fairly closely integrated into the historico-cultural and political traditions of the Muslim East. The rulers of the Arab Caliphate, the Golden Horde, the Ottoman Empire, Safavid Iran and Shirvan served as models for the region's elites, or — more figuratively — as "arbiters of political style." We shall also look at feudal formations that existed on the North-Eastern Caucasus before the emergence of the Shamkhalate, in particular at the history of the Khazar Khaganate.

It is well attested, then, that the mountain town of Kazi-Kumukh served as the shamkhals' residence in the summer months, while the coastal town of Tarki (Targhu) served them in winter. Such a dichotomy of residences was a very widespread phenomenon along the borderlands of the Great Steppe, which ran adjacent to — and densely intertwined with — the North-Eastern Caucasus.

Steppe precedents

Chinese chronicles from the Tang dynasty contain information that directly attests to the existence of two seats among Turkic rulers, explaining this by concern for their herds (Malyavkin 1989, p. 238). The Khazar qaghans probably inherited the custom of maintaining two residences from the qaghans of the Western Turkic Khaganate. The Arab geographer al-Istakhri notes that the Khazar qaghan lived in the city of Itil only in winter; for the whole summer he went out into the steppe, pasturing horses and growing wheat, rice and millet. Qaghan Joseph, in his famous letter, mentions the month of Nisan (April) as the date of his departure for the summer pastures (Polyud'e 2009, p. 293).

The city of Sarai al-Jadid (Sarai-Berke) was the winter residence of Uzbek Khan and was regarded as the capital of the entire Golden Horde. The summer residence was located at Beshtau (Ibid., p. 299). The summer residence of the Crimean khans, in turn, was the estate of Ashlama-Saray, while the winter residence was the capital of the Khanate — Bakhchisaray.

In every case cited above we see that it was the winter residence that held the status of capital. Here the rulers passed the time from October to April — that is, the most severe part of the year.

The polyudye and the shamkhal's tribute circuit

At the same time, the holding of two — or more — residences was not necessarily tied to the driving of herds and to a semi-nomadic way of life. Thus, the compilers of the collected volume Polyud'e — in particular Yu.M. Kobishchanov — have reasonably suggested that, in moving through his domain, the shamkhal was collecting tribute (Ibid., p. 266). The written sources confirm this (Kusheva 1963, p. 52). Evidence of the collection of payments in kind from the subject population by feudal lords — that is, of the Slavic practice of polyudye (a circuit of tribute collection) — is found also in folklore. One of the old Kumyk songs, for instance, mentions the Erpeli prince Tonay, who was killed while collecting dues from the inhabitants of the settlements of his domain (SMOMPK 17, 1893, pp. 51–52). The same kind of duty-collection is suggested by the "Register of dues to the shamkhal," compiled no later than the 1570s, which clearly demonstrates the payment of dues in kind rather than in money (Shikhsaidov 1977, pp. 108–109). Such dues were known among the Kumyks under the name yasak (Orazayev 2003, p. 190). Kobishchanov was perhaps the first to formulate this view of the traditional seasonal movements of Dagestani lords from one settlement to another (Kobishchanov 1989, pp. 42–43). Besides the shamkhals, one may cite the example of the rulers of Kaitag — the utsmis — who had residences in the mountains (Kala-Koreish), in the foothills (Majalis and Yangikent), and on the coast (Bashly).

Surkhay and the mid-17th-century shift

In the context of the dying out of the polyudye tradition, a request from Shamkhal Surkhay in the very middle of the 17th century is of interest. It is set out in a letter to Dugri Nutsal, ruler of the Avar principality of Avaria: "When the bearer of this letter — my mulazim Ahmad — comes to you, spare no effort in taking my kharaj from the Chamalal jamaat. Show no negligence in this matter" (Aytberov 1982). In the mid-17th century, then, the shamkhal was resorting to the help of an ally and to the services of a special functionary — designated by the word mulazim — to collect his kharaj. In our view it is no accident that Kazi-Kumukh's departure from the Shamkhalate took place precisely after the abandonment of the polyudye practice, that is, in the second half of the 17th century. Despite the fact that the word mulazim has an Arabic semantic base, in the period in question it was characteristic above all of the Safavid state, where it denoted a member of the ruler's household retinue. In our view, Surkhay — who had lived for a period at the shah's court and had kin among the influential Qizilbash nobility — was attempting to carry out reforms "in the Safavid spirit" within his own domain. One of these reforms was precisely the transfer of yasak collection to his retainers.

European parallels

Travelling through the towns of one's domain — and the absence of a single capital — was also characteristic of 15th-century Flanders, where the Dukes of Burgundy had residences in Ghent, Bruges and other cities (Polyud'e 2009, p. 105). In that case, however, the principal reason for such movements was above all the dukes' wish to prevent the erosion of their own sovereignty in a distant and wealthy province which had, besides, a centuries-long history of resistance to French and other rulers. The capital of Burgundy proper was Dijon, which lay a considerable distance from Flanders.

World history offers other examples of reasons that forced rulers to change residence periodically — in particular, under pressure from their subjects. Thus, for example, in March 1531 the estates of Saxony, alarmed by the heavy expenditure of the court of Elector Johann, advised him to live with his court at the residences of Weimar, Torgau and Coburg, alternating them in turn no more than a year or half a year at a time, "so as not to exhaust the grain reserves entirely, not to cut down all the surrounding forests, not to drain the ponds, and not to burden the people excessively with dues" (Tatsenko 1990, p. 121). Moreover, Saxony, right down to the end of the 16th century, did not even have a single fixed capital. Only the complication and growth of central administrative structures led to the emergence of a permanent seat of government.

After 1640: not two residences, but several

Among the key arguments of those who see Kazi-Kumukh as the principal residence is the cessation of the shamkhals' presence in Kazi-Kumukh after 1640, when the people of Kazi-Kumukh are said to have risen in revolt and driven out the shamkhals. Yet the sources show that Kazi-Kumukh was part of the Shamkhalate both in the 1650s and even in the 1680s (Shmelev 2000, pp. 92–93). At the same time, the shamkhals did indeed cease to move to Kazi-Kumukh for the summer. Evliya Çelebi and Adam Olearius, for instance — who left highly detailed accounts of the shamkhals and the Shamkhalate — do not report any such relocations. It should be particularly underlined that Adam Olearius visited Tarki in 1636 — that is, before the notorious "expulsion."

Like the utsmis of Kaitag, the Dukes of Burgundy and the Electors of Saxony, the shamkhals already in the 16th–17th centuries had at their disposal not two, but several residences. Among them were Boynak (now the village of Ulluby-Aul in the Karabudakhkent district of the Republic of Dagestan), Kafir-Kumukh and Kazanishche (now villages in the Buynaksk district of the Republic of Dagestan).

One of the founders of modern Dagestani historical scholarship, R.M. Magomedov, wrote: "That Kafir-Kumukh was for a long time the summer residence of the shamkhals admits of no doubt; on the contrary, there are every grounds for asserting that the original residence of the shamkhals was precisely this settlement" (Magomedov 1957, p. 146).

Moreover, the 17th-century Turkish author Evliya Çelebi, well informed about the history of the North Caucasus, reports — citing the author of the work Fatawa-i Tatarkhaniyah, Tatarkhan, son of the Delhi sultan Ghiyath al-Din Tughluq, who lived in the first half of the 14th century — that in antiquity the capital of the Dagestani padishahs (the shamkhals) was Djulat (Otryvki, l. 85). The same author applied, to the city of Endirey — a considerable one in his own day — the epithet "seat of the padishah of Dagestan" (Çelebi 1979, pp. 114–115).

Kazi-Kumukh's Simirdal and the link to Semender

A further point of interest, against those who see Kazi-Kumukh as the primary seat, is the presence at Kazi-Kumukh of the Simirdal cemetery. Its name, in the view of the ethnographer A.G. Bulatova, derives from that of the capital of the Khazar Khaganate — the city of Semender, which the majority of historians locate at the site of Tarki and modern Makhachkala. More significant still is the fact that the people of Kazi-Kumukh likewise used the word simirdal — that is, "the Semender folk" — for the shamkhals themselves and for their successors, the local khans (Bulatova 1999, p. 58). The city of Semender, in the view of the great majority of researchers, lay on the territory of the present-day village of Tarki and of modern Makhachkala (Gasanov 2015, p. 135) — that is, on the site of the shamkhals' winter residence. Such a coincidence cannot be considered accidental. To complete the picture, the inhabitants of Kazi-Kumukh also used, in reference to Tarki, the name Azayni — going back to the Arabic word ʿaẓami / ʿaẓīm (from which, incidentally, the word "azimuth" also derives), meaning "great" (Aliyev 2001, p. 45), in this context "principal / central, serving as a point of reference."

Conclusion

To sum up: the tradition of the shamkhals dwelling alternately in two or more residences is not an exceptional phenomenon; it has precedents in world history, the closest geographically being in the past of Khazaria and the Golden Horde. In both of those state formations, moreover, the status of capital was assigned precisely to the winter residence.