The Greeks May Have the Last Word but Who Has the First?

Elliott B. Martin, MD        

University of Connecticut, School of Medicine, Farmington, CT

The familiar medical roots ‘-iatry, -iatrics, iatro-,’ and their variants traditionally have traced their etymology to the Attic Greek word for physician, ‘iatros’. This paper traces the etymology of ‘iatros’ itself. Proceeding stepwise through time, the article demonstrates the evolution and borrowing of the word from its immediate Archaic Greek predecessor, the ‘iatār’, and further back to its earliest Greek form in the Linear B inscriptions. Beyond the Greek, it is then demonstrated how the Linear B was a direct borrowing from the non-Greek Linear A, an earlier language of the Ancient Mediterranean. From there, the article examines the likely cognate forms in the even earlier Hittite, Egyptian, and Akkadian languages to the East. Ultimately the origin of the ‘iatros’, of our English root ‘-iatry’, is traced to the earliest recorded language, Sumerian, and the Sumerian word for physician, the ‘IA.ZU’.

Keywords: ancient medicine, iatros, asu, ashipu, ia.zu, Mesopotamian medicine, Greek medicine, Egyptian medicine

For full text of this article, including supplemental notes please download the PDF at the end of the page.

Share on Facebook.

Writing is still the most formidable technological advancement of humankind. The manipulation of words has been both an art and a science since Mesopotamian scribes first laid stylus to clay. More importantly, the words themselves – both revered and reviled, misspelled and misspoken, modified and manipulated through generations and across cultures – offer insights like no other material artifact into the intellectual history of civilizations, both those long since passed, and our very own today. Precision of written language, based upon a common scholarly education, played an enormous role in the codification and pursuit of natural science from the Crusades through the Renaissance, and for more than a century after the Enlightenment.  But how many of us, raised now at least a generation or two removed from the decline and fall of ‘classical education’ – that is, education firmly rooted in the study of Greek and Latin – are aware even that the word ‘science’ comes to us most recently from the Latin root scientia?  Scientia at one time meant ‘knowledge’, a fact that may or may not be self-evident, but armed with that information, how many of us then wonder from where it is the Latin root itself emerged? The English language, after all, does not begin and end with Latin and Greek. Words existed for thousands of years before the time-worn epithets of Homer, and for even more years before Cicero turned his logic-bending courtroom jabs. Indeed, for all the intellectual doors opened by way of the classical education of our not-so-distant ancestors, many other doors have become unwittingly neglected. Caught now, for example, between ‘amyotrophic lateral sclerosis’ on the one hand, and ‘Lou Gehrig’s disease’ on the other, where does that leave the current generation of scientists? Fortunately, thanks again to writing, we still have the words themselves as clues. The twenty-four hundred year-old oath of Hippocrates suggests, if not mandates, that certain moral and ethical standards remain, have remained, and will remain constant over time. Words is words is words, so to speak, and to offer some perspective, it is appropriate to remember that Hippocrates’ oath was already four hundred years old by the time of the historical Christ. On the streets of Nazareth and Jerusalem Christ likely spoke the koine, or common Greek of his time, and how might a carpenter’s son, a world away from classical Greece, have understood the, even by then, archaic phraseology of his own scientist forefathers? To our perspective, Shakespeare had just completed The Tempest four hundred years ago.

The utility of the study of Latin and Greek as prerequisite for a more enriched medical educational experience was the subject of much debate early in the twentieth century.  The controversy then, however, was not whether undergraduate students should study these languages at all, but whether they should study them for four to six years versus the more standard two. In defense, however, of the current, often grammar-less, abbreviating, emailing, text-messaging generation, there does exist an even more distant past from which physicians draw everyday, a past that predates Hippocrates by two thousand years, that predates Galen by twenty-six hundred years, and to which both they and we owe more of our medical tradition than is currently held. Whenever we speak the words unique to the field of medicine we are speaking the most conserved, the most preserved words of the English language. In other words, we are speaking not only Latin and ancient Greek, but the Hebrew, Aramaic, and Canaanite of the ancient Levant; the Hittite of ancient Troy. We are speaking the Egyptian of the Great Pyramids. We are speaking Akkadian, the language of ancient Babylon and Assyria, and ultimately we are yet speaking Sumerian, the very first recorded language. We are in fact speaking languages as yet undeciphered, and likely as yet undiscovered. For although much has been made of the Greek and Latin origins of many of our current medical terms, little has thus far been said of the Near Eastern origins of the Greek and Latin terms themselves. This paper will present one example, a single word, synonymous with physician, that has survived, in various incarnations, the passing of not only centuries, but millennia: the ancient Greek word iatros. Literally ‘healer’, the iatros is whence derives our previously well understood loans from the Greeks, ‘-iatry’, ‘-iatrist’, ‘-iatrics’, ‘iatro-’, et cetera.

Let us begin then where many scholars have traditionally begun (and ended), with the exhaustively studied ‘Golden Age’ of Ancient Greece, the fifth century B.C. This era was the time of Hippocrates himself, wherein the standard Attic Greek word for physician, for healer, had become, and would remain, iatros (ya-tros).  Traditionally, etymologists have stopped there.  This was after all the fabled Periclean Age, the age of Greek empire, the time of the great playwrights. This was, in other words, the highest period of the ancient Greek language. If, however, we take a further step backwards and look at the Greek language three hundred years before the Golden Age, during what is known as the Archaic period of ancient Greece, we see that the word for physician, for healer, was iatār (ya-tār). This is the same word as iatros only with a different, older case ending. This would have been the time of Homer and Hesiod, the time of the lyric poet, Sappho. Three hundred years is a long time (if one considers how much the English language has changed since the early 1700’s), but in fact, this word iatār for physician appears as well within the earliest known written Greek corpus, the Linear B inscriptions of the second millennium B.C.  These administrative records of the great palace civilizations of Mycenae, Knossos, and Thebes predate Homer by about seven hundred years, taking us back in time now to as early as the fifteenth century B.C., a good thousand years or so before Plato was even born (consider the state of the English language in the early 1000’s). These ancient city-states are those very kingdoms that inspired the epic heroes of Homer – Achilles, Agamemnon, Odysseus – as well as the legendary iatār of all those Greek forces encamped outside the walls of Troy, Paean himself. 

The iatār can be traced even further back than this, however, and without ever leaving the Aegean. He can be found in the Minoan civilization of ancient Crete – the civilization of King Minos, his daughter Ariadne, and his monstrous illegitimate son, the Minotaur. The language of the Minoans, or at least the brief written record of it that survives, has become known as Linear A, having been discovered slightly earlier than the equally colorlessly named Linear B,.  Although Linear A has yet to be definitively deciphered, its writing system shares nearly two-thirds of the same pictographic and ideographic characters as Linear B, indicating that Linear B most likely borrowed its system from the earlier Linear A.  But whereas Linear B is most definitely Greek, Linear A is most likely not. What can we possibly learn then from a secret code that has yet to be cracked?  As far as our present discussion is concerned we can learn much. Given the orthographic, or handwriting, congruity between Linears A and B, the fact that Linear B has been fully deciphered, and the geographical proximity between the two languages, it is fairly certain that certain repeated combinations of signs common to both languages are shared words. Luckily for our present discussion, one of these commonalities is the word iatār, a term that appears syllabically in both Linears A and B as i-ya-te.  I-ya-te  means ‘physician’ in Linear B – by analogy the same in Linear A – which would put the earliest Aegean reference to the iatār, or physician, perhaps as early as 1850 B.C.  This is fourteen hundred years before the time of Hippocrates, and either earlier than or contemporary with the great Babylonian king, Hammurapi.  The presence of the iatār, or i-ya-te, in Linear A, a non-Greek language older than written Greek, also suggests quite strongly that the word is not Greek in origin.

And with that we can no longer accept that our very own ‘-iatrists’ and ‘-iatries’ owe their lineage solely to the iatros, the iatār. As stated above, there are languages older than Greek, and from where then did this particular father of modern medicine ultimately emerge? Which ancient civilization produced this most ancient of healers? Although the Greeks certainly advanced the art and science of the iatār, they did not in fact produce the iatār. Like ours, theirs, too, was a long-standing tradition of medicine. Like ours, theirs, too, was a vernacular enriched by the distillation of centuries and millennia. And Linear A remains, for a stretch of centuries, the only extant thread of evidence of literacy west of the Near East. To the East then lies our next logical place of inquiry.

The nearest empire contemporaneous with the Minoans was that of the Hittites, who occupied Anatolia, modern day Turkey, throughout the middle of the second millennium B.C. The Hittites, like the Minoans and many of the Mycenaean city-states, eventually fell to the fabled Sea Peoples, themselves the likely marauding Greek heroes upon whom the Homeric tales were based. (The city of Troy was a coastal satellite of the Hittites, and even the great historian Thucydides brands his storied ancestors as little more than pirates.) The Hittites appear prominently in the Bible as well , and their language stands as a sort of orthographic, if not exactly linguistic, bridge between the East and the West. Hittite scholars not only borrowed the abstracted cuneiform writing system of the Babylonians and Assyrians to the East, but they created their own hieroglyphs as well, based on Egyptian, Cypriot, and Linear A pictographs to the West. What is perhaps most amazing about this is that Hittite is related, linguistically, to none of these other languages. The Akkadian language common to both Babylon and Assyria is Semitic; Egyptian is Hamitic, with strong ancestral ties to Akkadian; and Cypriot and Linear A, though both remain undeciphered, have both withstood scholarly attempts at any substantial Hittite connection.  Hittite is, however, related to Greek and Latin, all three of which are of Indo-European origin. So, does our iatār then come from some common Indo-European root?

Sanskrit, the first written tongue of the Indus Valley civilizations, is the oldest ancestor we have of the Indo-European languages . There are three main terms in Sanskrit for physician. First is the ayurvaidya, or ‘the one who knows the duration of life’ , second – and commonest – is the reduplicated cikitsaka, or ‘physician’ , and third is the bhits aya, meaning something like ‘healer’ , but which etymology is most mysterious. Separated by only a century or two, one would expect something similar to the Sanskrit in Hittite. What we find, rather, is that the word for physician in Hittite appears in texts exclusively as a Sumerian logogram. That is, the written Hittite word for physician is a direct borrowing from the Sumerian written representation of physician, the grouping of three consecutive cuneiform signs LU2.A.ZU.  In Sumerian, the LU2 sign in such context (otherwise literally meaning ‘man’) was not read, but designated a category, in this case an occupation.  The a-zu was a physician. Traditional convention holds that the Sumerogram masks a proper Hittite word, and that the Sumerian was not meant to be read as such. It is obvious by the mere spelling of the word that the Hittites indeed borrowed a Sumerogram representation of physician, but probably from the Akkadian language, which stands geographically between Sumerian and Hittite.  This borrowing occurs often in Hittite, as it does in Akkadian. The scholarly consensus with regard to the reading of such foreign words is that they should be read in the native language of the text. In other words, lu2-a-zu should be read not as a-zu (its Sumerian reading) but presumably as some variation of ayurvaidya, or some such. But to accept this reading would be to accept that the Hittites would have borrowed a foreign spelling for a native word. The argument for reading foreign words as native in Near Eastern cuneiform stems from the facts that many Sumerograms and Akkadograms are appended by case endings or other suffices in the native language, and that the native word also appears with some frequency in its native writing.  The problem in this case is that the Hittite word for physician only ever appears in Sumerian, at least in surviving texts thus far. What is most likely then is that the Hittites simply borrowed the word from their Akkadian neighbors and conquerors. In other words, the Hittite word for physician was probably the Akkadian word for physician. And coincidentally enough, the Akkadian word for physician is a-tsu, borrowed directly from the Sumerians.

Turning then to Babylon and Assyria, the two great Mesopotamian empires that emerged from what is now modern day Iraq, tied together culturally by their common Akkadian language – the common ancestor of all Semitic languages – we find the Akkadian physician known as the asu, likely pronounced ‘ah-tsu’.  Here at last we have what appears a likely candidate for direct descendant of our Linear A i-ya-te, and thus grand-descendant of our Greek iatār. The asu, after all, was practicing medicine from at least the third millennium B.C., within an empire that stretched well into the Aegean at about the time of the appearance of Linear A.  As for the apparently free interchange of ‘z’, ‘s’, and ‘t’, the pronunciation of sibilants in ancient languages has been the subject of much unresolved debate.  For our study, it is sufficient to note that it is not uncommon for a sibilant in one language, ancient or otherwise, to be replaced by a dental stop in another, i.e., ‘t’ for ‘s’.

From where then does the Akkadian asu come?  This takes us to the latter half of the third millennium B.C. As noted above, the Babylonian asu is itself a loan word from the even earlier Sumerian a-zu.  Historically known as ‘the birthplace of civilization’, Sumer was the Biblical birthplace of Abraham and the site of the mythological Tower of Babel. It is the place, appropriately enough, where writing was invented. What is most interesting about Sumerian is that it is a language isolate, meaning that it is unrelated to any known language, ancient or otherwise.  We have only been able to reconstruct what we do know of Sumerian thanks to the Akkadian reliance on Sumerian pictograms and ideograms and the survival of Sumerian-Akkadian bilingual sign-lists and lexical lists.  (Sumerian was the scholarly language of Babylonian scribes, not unlike the academic use of Latin throughout the European Middle Ages, Renaissance, and Enlightenment.) From these it can be gathered that the Sumerians had two types of physicians, the ia-zu and the a-zu, literally the ‘oil healer’ and the ‘water healer’, respectively (ia is ‘oil’, and a is ‘water’ in Sumerian; zu typically is the verb ‘to know’. But given the variant spellings of zu it is not certain that a-zu should be considered literally as ‘knower of water’ ). The difference between the two, if indeed there is any, is not well understood.  What is clear is that the ia-zu and a-zu were the physicians of ancient Sumer, the most ancient of physicians in the written record, and between them we have at last found the most probable direct and most ancient descendants of the Akkadian asu and the Linear A i-ya-te. In plain English, we have at last found the likely great-grandparent of our Greek iatār, the great-great-grandparent of our modern ‘-iatrics’ and ‘-iatries’.

All this is only an example of how it can be demonstrated that the very language itself of our medical terminology can be traced back continuously to the very dawn of recorded history. Every time one speaks and writes the words of the exclusive medical vernacular one is speaking and writing not only English, not only Latin and Greek even; one is speaking Egyptian, Hittite, and Sanskrit. One is speaking Akkadian and Sumerian. One is speaking remnants of not-so-lost languages. Medicine has always been boastful of and humbled by its well-known heritage with the ancient cultures of Rome and Greece. But just as these two great empires owed much to their predecessors, so, too, are those of us descended from Greco-Roman tradition, indebted.

Elliott B. Martin, Jr. is a first-year resident in Psychiatry at the University of Connecticut Health Center. He also has a Master’s in Near Eastern Languages and Cultures from UCLA, and is formally advanced to doctoral candidacy. A former instructor in both Classics and NELC, and a former secondary school teacher, medicine is a second career for Dr. Martin. He can be reached at:

Historia Medicinae Editor,
Dec 31, 2009, 10:18 AM