Speaking of Words
By MICHAEL FERBER
The Hittites were once a powerful kingdom that, at its peak around 1300 BCE, dominated Anatolia (now Turkey) and even reached as far south as Aleppo (now in modern Syria); they fought a major battle with the Egyptians, led by Ramesses the Great, at Kadesh in 1274. They didn’t call themselves Hittites but the people of Nesha, which was once their capital, and they called their language Neshili. As they grew stronger, however, they moved their capital to Hattushas, named after the earlier Hatti people, and those names gave rise to the Hebrew term Hitti (or Chitti), and then our word for them.
By the late twelfth century, battered by Assyrians to the east and Phrygians to the west, the Hittite empire broke up; by the eighth century the last remnants were absorbed by neighboring states and their language disappeared.
But in several Hittite cities troves of clay tablets have come to light, with writing on them dating from about 1700 to 1300 BCE, in cuneiform, the wedge-shaped characters invented by the Sumerians and used by the Babylonians and other peoples not related to the Hittites. But deciphering the tablets proved difficult, even though the phonetic values of some of the symbols could be guessed at from their uses in other languages. Other symbols seemed to be logograms, which were not sounds but meanings, such as “city” or “king.” After many efforts had failed, however, the Czech linguist Bedrich Hrozny noticed a passage that had the logogram for “bread” and the phonetic word wa-a-tar. Near the sign for “bread” was ezz-a-te-ni. Surely waatar meant “water” (like Greek hudor, earlier wudor, and of course our water), and ezzateni meant “eat” (like German essen, Latin edo, and our eat). In other texts the word for “wine” is wiyana. The Hittites, it turned out, were our linguistic kin!
Though Hrozny was not the first to think Hittite belonged to the Indo-European family, his work was decisive. Since 1915, when he published his first paper, a great deal of work has been done, and now we know quite a bit about its grammar, vocabulary, and word order. We have also learned about the Hittites’ gods and ceremonies, their laws, and their diplomatic relations with other countries. There are even references to a city called Wilusha, which may be the one the Greeks called Ilios, earlier Wilios, and to a place nearby (or the same place) called Truwisha, and to a king of Wilusha named Alaksandus, who made a treaty with the Hittite king in 1280 BCE. Alaksandus sounds a lot like Alexandros, whose other name is Paris, the prince of Troy-Truwisha and seducer of Helen in Homer’s Iliad.
All this is wonderful to learn about. But the language is very odd compared to the other Indo-European languages. Its verbs have only two tenses, present and past. They have only two moods, indicative and imperative; they have no subjunctive, unlike Latin, Greek, Vedic Sanskrit, and English; and no optative, unlike Greek and Sanskrit. They have only one aspect, which means they do not distinguish between “I eat bread,” “I am eating bread,” and “I have eaten bread,” though there are apparently some work-arounds. It all seems surprisingly primitive.
In contrast to the skimpy verb system, however, nouns come in quite a few cases (like Sanskrit), but in only two genders, “common” and neuter; there is no distinction between feminine and masculine genders, as we find in Greek, Latin, Sanskrit, Russian, German, and Old English. One of the cases, the “ergative” or “agential” case, is unique among the other ancient sister languages. It seems that, sometime in the pre-history of these languages, neuter nouns, which usually meant things not alive or active, such as “stone” or “wall,” could not be subjects of active verbs. They could not be agents. When the Hittites wanted to make certain neuter nouns into agents, they recuited an ending or suffix that had only attached to “common” nouns, which could be agents, and attached it to the neuter ones. Got that? At any rate, Hittite is weird.
So what happened to make Hittite so different? There has been a contentious debate about this since its decipherment. Both sides agree that the ancestor of Hittite and the other Anatolian languages (probably including what the Trojans spoke) split off from the other Indo-European languages very early, but one group of linguists thinks Hittite then lost the elaborate verb system of the mother tongue while the other group thinks it preserved the simple system of the mother tongue while the other daughters innovated and elaborated it. Amidst this debate another one has intermittently popped up: Is Hittite a daughter of Proto-Indo-European or a sister, or, to be strict, a niece? Was there a Proto-Anatolian of equal status to PIE? This is partly just a matter of nomenclature, but it might affect what assumptions linguists make as they reconstruct the mother tongue.
It will certainly affect our ideas concerning which people first spoke something we call PIE and what people first spoke PA, and, if the tongues were sisters, which people spoke their common ancestor, Proto-Indo-Anatolian. And that’s the lastest news. The answer may be found in the decoding of some ancient human DNA.
The consensus now is that the first PIE speakers were those whom archeologists call the Yamnaya, a pastoral people who lived in the Pontic-Caspian steppe (now Ukraine and Russia) between about 4000 and 3200 BCE, at which point one group migrated east into what is now China and millennia later wrote some documents in what we call Tocharian. The Yamnaya built great burial mounds called kurgans for their lords. In another 500 years or so they learned to ride horses and attach them to wagons and chariots, and soon they invaded Persia, India as far as Nepal, and Europe as far as Ireland.
Now if Hittite were a daughter of the language of the Yamnaya, we would expect to find Yamnaya genetic markers in the DNA of ancient Hittites. It would imply that some descendents of the Yamnaya migrated south to Anatolia, either by the western route through the Balkans, as the Proto-Greeks did, or by the eastern route through the Caucasus. This has also been a contentious topic, but now it is rendered moot. Ancient DNA analyses show virtually no Yamnaya ancestry in Anatolia.
So where did the Proto-Anatolians come from? The Caucasus, and that’s where the Yamnaya came from too. Both groups blended with existing inhabitants, and no doubt their languages blended too, at least a little. And who were their ancestors? A major study from the Harvard ancient DNA lab, in a paper published in 2025, is that a group of people living in the Caucasus and Lower Volga region around 4400 BCE have been identified as having a few distinctive genetic markers; these CLV markers are found abundantly in the DNA of the Yamnaya people and less plentifully but still distinctively in the DNA of the Hittites. That would seem to settle it. The CLV people must have spoken Proto-Indo-Anatolian, though I doubt if they called it that. Sometime after 4400 many of them moved north into the steppe and somewhat fewer of them moved south into Anatolia.
So that’s the latest news. Future news may complicate all this, of course. But if you’ve made it this far, pour yourself a glass of wiyana, or at least a glass of sparkling waatar, and toast your Anatolian cousins.
I am happy to hear from readers with questions or comments: mferber@unh.edu.
Michael Ferber moved to New Hampshire in 1987 to join the English Department at UNH, from which he is now retired. Before that he earned his BA in Ancient Greek at Swarthmore College and his doctorate in English at Harvard, taught at Yale, and served on the staff of the Coalition for a New Foreign Policy in Washington, DC. In 1968 he stood trial in Federal Court in Boston for conspiracy to violate the draft law, with the pediatrician Benjamin Spock and three other men. He has published many books and articles on literature, and has a deep interest in linguistics. He is married to Susan Arnold; they have a daughter in San Francisco