The Caspian Sea

The Greek name of the Caspian, Kaspía Thálassa, after the Káspioi, the Caspii, a people of its southwestern shore. Herodotus, remarkably, gets the sea right where later writers got it wrong: he states plainly that the Caspian is “a sea by itself, having no connection with any other,” enclosed on every side, against the common Greek view, taken from Alexander’s surveyors, that it was a gulf of the encircling northern Ocean. Beside Kaspía the Greeks also used Hyrkanía Thálassa, the “Hyrcanian Sea,” after the rich Iranian province of Hyrcania on its southern coast.

The two Greek names are the first instances of the sea’s habit of being renamed for whatever people held its shore. Kaspía fixed on the Caspii of the southwest, Hyrkanía on the Hyrcanians of the south, and the same impulse would later give the Arabs their Baḥr al-Khazar and Baḥr Jurjān. The sea too large to belong to anyone became a sea named, in rotation, for everyone, and the Greeks, alone unsure even whether it was open or closed, started the rotation with two names at once.

The Latin name of the sea, Mare Caspium, taken from the Greek, with the alternative Mare Hyrcanum used just as the Greeks used both names. Pliny inherits and repeats the old error that it opens by a narrow channel into the northern Ocean, a mistake that persisted in European geography for centuries; the Romans never reached it closely enough to correct what Alexander’s surveyors had reported.

The Latin carried Caspium and Hyrcanum into the European learned tradition, and it is the Caspian of the Caspii, the southwestern people, that finally prevailed in the modern languages over the Hyrcanian of the southern province. Of all the names the sea collected, shore by shore and empire by empire, the one that lasted was the Greeks’ first choice, a small people on one corner of a sea so vast that no one could see all of it at once.

The Armenian name of the sea, Kaspicʿ cov, “sea of the Caspians”: Kaspicʿ is the genitive plural of Kaspkʿ, the Armenian name of the Caspii, so the phrase means exactly what the Greek Kaspía Thálassa means, built instead from Armenian’s own grammar. The name appears in the Ašxarhacʿoycʿ, the seventh-century Geography attributed to Anania Širakacʿi, which surveys the sea and the lands around it in the fullest geographical work to survive from early medieval Armenia. For Armenians the Caspii were not a people known only from books: their old country near the mouths of the Kura and Araxes lay within the historical province of Pʿaytakaran, where Kaspkʿ served as a district name.

Every other name in the kaspia family reached its language through the Greek geographic tradition; the Armenian form alone did not need the loan. Armenia bordered the Caspians’ country, absorbed it as a province, and inflected their name with its own genitive plural, so that the sea is named “of the Caspians” in living grammar rather than by an inherited adjective. The Caspii themselves disappeared from history in antiquity, yet modern Armenian still says Kaspicʿ cov: of all the languages on this page, the one that knew the eponymous people as neighbors is the one that has never stopped declining their name.

The medieval Latin continuation of Mare Caspium, carrying the classical name and the classical error together. Isidore of Seville, whose Etymologiae transmitted ancient geography to the Middle Ages, lists Caspius among the names the world-encircling Ocean takes from neighboring regions: for Isidore the Caspian is not a sea beside the Ocean but the Ocean itself, wearing a local name. The mappae mundi drew it accordingly, a gulf opening from the encircling waters, and seven centuries of Latin geography repeated the picture. Then the Franciscan William of Rubruck, sent across the steppe to the Mongol court in 1253, passed along the sea’s northern side in person and wrote the correction plainly: it is not true what Isidore says, for the sea nowhere touches the Ocean, but is on every side surrounded by land.

Herodotus had said the same thing seventeen centuries earlier, and the report of Alexander’s surveyors had overruled him; Isidore inherited the surveyors’ gulf, and Rubruck’s own eyes overruled Isidore. The name Mare Caspium held steady through the whole dispute, while the thing it named changed shape twice, from enclosed sea to gulf of Ocean and back. No entry on this page shows more plainly that a name can be far more stable than the geography behind it: the medieval West knew exactly what the sea was called through centuries in which nobody could say what it was.

The Arabic name of the Caspian, Baḥr al-Khazar, “the Sea of the Khazars,” after the Turkic people whose powerful khaganate dominated its northern steppe and the lower Volga through the early Islamic centuries. Beside it the geographers use Baḥr Jurjān, their form of Hyrcania, and Baḥr Ṭabaristān after the Iranian province on its southern shore, three names for one sea.

The Arabic naming continues exactly the pattern the Greeks had begun: the sea labeled, in each age, for whatever people held the stretch of coast the namers knew best. As the Caspii had given the Greeks Kaspía and the Hyrcanians Hyrkanía, so the Khazars now gave the Arabs Baḥr al-Khazar. When the Khazar khaganate fell, its name faded with it, and the sea reverted in the wider world to the oldest of its labels, the Caspian; but for the medieval Arab geographers it was, above all, the Khazars’ sea.

The Middle Persian name Kamrōd comes from the Bundahišn, the Pahlavi compendium of Zoroastrian cosmology redacted in the ninth century, whose chapter on the seas inventories the waters of creation: the world-ocean Frāxkard at the edge of the earth, and beside it the principal salt seas, among them the Pūdīg, which alone has tides, and the Kamrōd, “that which they pass by, in the north, in Taparistān.” A sea lying north of Tabaristan, the Iranian province on the Caspian’s southern coast, can hardly be any other water, and the identification with the Caspian has been standard since West’s edition; the name itself carries no ethnonym and no certain etymology, and it is that single locating clause that ties it to the map.

Every other name on this page was coined by geographers, naming the sea for whoever or whatever stood on its rim. Kamrōd alone comes from scripture: the sea appears not as a place observed but as an item in the inventory of creation, listed among cosmic waters in a book about how the world was made. Iranian speakers held the Caspian’s southern shore for the whole of its recorded antiquity, and they are precisely the namers who did not call it after a shore-people; the outsiders named the sea for its inhabitants, while the inhabitants’ own tradition placed it, without an eponym, in the order of the cosmos.

The Persian name of the sea, Daryā-yi Khazarān, “the Sea of the Khazars,” with the native daryā where the Arab geographers write baḥr. The Ḥudūd al-ʿĀlam, the anonymous geography of 982 composed for a prince of Gūzgānān and the oldest surviving work of Persian geographical prose, describes the sea under this name: the Ghūz to its east, the Khazars to its north, and the Iranian coastlands of Gīlān, Daylamān, Ṭabaristān, and Gurgān along its south. The Persian usage runs parallel to the Arabic Baḥr al-Khazar, both naming the sea for the Turkic khaganate that dominated its northern steppe through the early Islamic centuries.

The Khazar khaganate collapsed within a century of the Ḥudūd al-ʿĀlam, and elsewhere the sea’s naming rotation simply moved on, as it always had, to the next people or the next town on the rim. In Persian the rotation stopped. The name outlived the khaganate, the Mongol invasions, and the entire subsequent turnover of the sea’s shores, and daryā-yi Khazar remains the everyday Iranian name of the Caspian today. The language of the sea’s southern coast, with a dozen shore-names of its own to choose from, kept the name of a vanished steppe empire for a thousand years after the last Khazar was gone.

The Syriac name of the sea, yammā d-Qazbīn, “the Sea of Qazvin,” named for the Iranian city that lies inland of its southwestern corner, behind the Elburz mountains. The name belongs to the late Syriac geographical tradition, above all to Bar ʿEbroyo, the thirteenth-century Syriac Orthodox maphrian whose learned compendia drew the marine geography of the Syriac schools together with that of the Arabic geographers; it mirrors the Baḥr Qazwīn that appears among the sea’s many names in the Arabic tradition. The older Aramaic world had lain out of sight of the Caspian, and Syriac inherited no name for it to continue.

For the rivers and lands of its Mesopotamian home, Syriac reaches back through Aramaic to names older than the alphabet, Bethnahrin, Deqlath, Perāt; for the Caspian it had nothing, and when its scholars finally needed a name they adopted their neighbors’ newest convention, naming the sea for a famous city near the coast. Every other name on this page points at the water from its rim, a people or a province on the shore; this one points at a city from which the sea cannot even be seen. It is the label of scholars working at second hand, and honest about it: the sea named from a library, by the landmark on the road that led there.