What’s the Oldest Hebrew Inscription? A Reply to Christopher Rollston

<< Back to Three Takes on the Oldest Hebrew Inscription

In the May/June 2012 BAR, Christopher A. Rollston’s “What’s the Oldest Hebrew Inscription?” considered four contenders as candidates for the oldest Hebrew inscription. The distinguished senior Israeli epigrapher Aaron Demsky disagreed with Rollston’s conclusions. In this BAR web exclusive, Professor Demsky argues that two of the four contenders are Hebrew inscriptions – the Gezer Calendar and the Izbet Sartah Abecedary. Read Demsky’s response below and visit the BAS scholar’s study page Three Takes on the Oldest Hebrew Inscription for an additional response by renowned archaeologist and Khirbet Qeiyafa excavation director Yosef Garfinkel as well as Rollston’s original article.


What’s the Oldest Hebrew Inscription?–A Reply to Christopher Rollston

By Aaron Demsky

Aaron Demsky

Professor Christopher Rollston presents a thoughtful discussion of the oldest Hebrew inscription (BAR, May/June 2012), presenting four possible contenders for this honor (the Khirbet Qeiyafa Ostracon, the Gezer Calendar, the Izbet Sartah abecedary and the Tel Zayit abecedary).

He emphasizes two basic ways of determining a Hebrew document: first, that it was written in what is called a national Hebrew script, which, however, only emerged in the ninth century B.C.E.; and, second, that its language is Hebrew, spoken from an earlier period than the emergence of the Hebrew national script, as witnessed by certain early Biblical passages. These, he says, are the real criteria for determining what a Hebrew text is and from there we are able to answer the question, What is the oldest Hebrew inscription?

On the basis of this rigid standard, none of the four inscriptions, in Rollston’s estimation, meet these criteria. Even the latest—the Tel Zayit abecedary is, according to Rollston’s paleographic analysis “pure Phoenician,” though found in a Judahite context!

“Another factor,” he writes, “that can be important in determining the language of an inscription is provenance: Where did the inscription come from?” However, he qualifies this additional factor by noting that “provenance should normally not be the sole means of identifying what language a text is written in.”

I can relate to this argument for it was the late Professor Moshe Kochavi and I who called the Izbet Sartah abecedary the earliest Hebrew inscription, a statement for which we were chided by several scholars. Kochavi, a senior archaeologist excavating the whole area around Tel Aphek, emphasized that Izbet Sartah was an early Israelite border settlement facing the more developed Canaanite metropolis. This provenance was central to our argument. My colleagues Israel Finkelstein and Moshe Garsiel went even further and identified the site of Izbet Sartah as Biblical Eben-ezer (1 Samuel 4:1).

Interested in ancient inscriptions? Read Alan Millard’s assessment of the oldest alphabetic inscription ever found in Jerusalem in “Precursor to Paleo-Hebrew Script Discovered in Jerusalem.”

While I certainly agree with Professor Rollston that script and language would be the definite means of defining the cultural/ethnic background of the scribe and his audience, I ask, What do we do with epigrapha that predate the ninth century B.C.E. (before the emergence of a Hebrew national script)? I would therefore add some external factors such as the above-mentioned provenance, as well as the archaeological horizon of the site. In particular, I would note other distinctive features that an epigrapher might find in the text such as onomastics (names) and literary/scribal traditions.

A case in point is the Gezer Calendar (one of Rollston’s candidates), which is written in a southern Canaanite dialect shared by Phoenicians, Philistines (see the Ekron inscription*), and notably by (some of) the Northern Israelite tribes, as witnessed by the Samaria ostraca. These closely related dialects—not including Hebrew as spoken in Judah—share the linguistic feature of the contraction of the diphthong (two vowels like ai usually pronounced ay). In these dialects the diphthong is compressed or contracted so that it sounds like a long e. For instance, in the Gezer Calendar, we find the words qetz, “summer,” and possibly kel, “measuring,” whereas in Biblical Hebrew as spoken in Judah, and as in modern Hebrew, these words are pronounced qayitz or kayil respectively. This linguistic trait found in the Gezer calendar appears also in the Samaria ostraca from the Northern Kingdom of Israel. There the word yen (spelled yod-nun) meaning “wine” is different from the Judean form yayin (spelled yod-yo-nun) This linguistic fact shared by the Gezer Calendar and the northern Hebrew Samaria ostraca is significant in light of the fact that Gezer of the late tenth century, when this calendar is dated, was probably settled by the northern tribe of Ephraim (Joshua 21, 21; 1 Chronicles 7, 28; see also 1 Kings 9:15–17, where it is related that Gezer was given to King Solomon as a wedding gift by Pharaoh and came under Israelite control, and most probably resettlement).

The real clincher, however, is the generally recognized feature that the Gezer Calendar was signed by the copyist bearing the telltale name ABY, read by many scholars as a short form of ABY[AHU]. If so, it is a so-called Yahwistic name, translated “My father is YHW,” with the YHW element indicating the national deity of Israel. I would estimate that at this time, 99 percent of people with a Yahwistic name were Israelite. (Joseph Naveh, who considered the Gezer Calendar to be Phoenician, had to read the appendix [ABY] not as a personal name but as another writing exercise, an abbreviated abecedary including the first three letters of the Hebrew alphabet—ABG!) It seems to me that considering the generally accepted signature ABY[AHU] as well as the historic context of the settlement pattern of the site and leaving aside the ethnically indistinguishable dialect and script shared with its neighbors, we can say with some degree of certainty that the Gezer Calendar was written by and for an Israelite audience. In a broad sense it is a Hebrew inscription.

In the enhanced edition of the BAS Ancient Inscriptions: Voices from the Biblical World CD-ROM, you can view 295 maps, color images and drawings, accompanied by captions written by Professor P. Kyle McCarter of Johns Hopkins University. Ancient Inscriptions: Voices from the Biblical World is a unique tool for scholars, teachers and laypeople who are fascinated by the ultimate human artifact: the words of our ancient ancestors.

I must take exception to Rollston’s statement regarding abecedaries, especially when he refers to the Izbet Sartah inscription and says “It is also the easiest to dispose of if the only question is whether it is a Hebrew inscription. Since it is an abecedary, we can ask only whether the script is Old Hebrew.” Of course, if we were dealing here with the standard order of the letters of the alphabet, we would not be able to say much about the novice writing it down. However, if the abecedary preserves an alternate tradition of the letter order, as do the Izbet Sartah and Tel Zayit abecedaries, then we are looking at different scribal traditions in ancient Israel, which then become a cultural marker with historic implications.

When I deciphered the Izbet Sartah inscription in 1976,** I proposed that the abecedary with the four transposed letters het/zain and peh/‘ain represented an alternate order of the Canaanite alphabet. I supported this theory by noting that the peh/‘ain order was found hundreds of years later in the alphabetic acrostics in the Book of Lamentations, chapters 2–4. Happily, this letter order was soon found in the Hebrew graffiti from Kuntillet Ajrud c. 800 B.C.E.*** More recently it and other transposed letters, including the het/zain order, were found in another Judean context in Tel Zayit c. 900 B.C.E. I maintain that abecedaries cannot easily be dismissed or considered for paleographic analysis alone, especially if they are variables of the standard order. Until we find another example of an abecedary with these transpositions of letters from Phoenicia, Aram or Greece, we must conclude that these variants point to a particularly innovative Judahite/Israelite scribal tradition. Furthermore, since this elementary scribal tradition is also found in Biblical texts (Psalms and Lamentations), we must say that Izbet Sartah is the earliest evidence of a long literary tradition peculiar to ancient Israel.

In conclusion, there are various distinctive means, in addition to script and language, that aid in determining the language or cultural background of a document written prior to the ninth century B.C.E. For one, the provenance and what it says about the contemporary settlement pattern, in addition to the archaeological horizon in which the text was found would give us an idea of the writer’s identity and that of his readership. Moreover, onomastic details as well as traces of a literary tradition would provide answers regarding the cultural and possible religious background of the scribe that are not provided by the ethnically indistinctive script and unheard language. If that is so, then the Izbet Sartah ostracon must be considered to be the oldest inscription emanating from a Hebrew cultural milieu.

In the September/October BAR, Yosef Garfinkel wrote an additional response critiquing Rollston’s study. Read Christopher Rollston’s “What’s the Oldest Hebrew Inscription?” and Garfinkel’s “Christopher Rollston’s Methodology of Caution” in the online Scholar’s Study discussion “Three Scholars Take on the Oldest Hebrew Inscription.”



* Aaron Demsky, “Discovering a Goddess: A New Look at the Ekron Inscription Identifies a Mysterious Deity,” BAR, September/October 1998.

** See Aaron Demsky and Moshe Kochavi, “An Alphabet from the Days of the Judges,” BAR, September/October 1978.

*** Mitchell First, “Can Archaeology Help Date the Psalms?” BAR, July/August 2012.

Posted in Inscriptions.

Tagged with , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , .

Add Your Comments

3 Responses

Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.

  • David says

    I would love to see where the Temple was FIVE times taller than the pyramids. Maybe if you add the height of Mount Moriah above sea level. Otherwise, not remotely possible or documented.

  • Joseph says

    philippe says.

    There are good reasons to challenge the premise the Hebrew is a derivative of Phoenician, Sumer, Canaanite, etc.

    1. Please produce the equivalent of Hebrew in any previous nation’s language, especially one older & mightier than Israel? No excuses can impact here: the Hebrews were constantly in exile, dispersion and wars; others were not so encumbered. Where then is the Phoenician Bible or its alphabetical books as seen in the Hebrew?

    2. The Phoenicians were allies of Israel and conducted joint naval brigades with King Solomon, by which time a host of advanced Hebrew alphabetical books were already completed [The Torah, Book of Joshua, Book of Judges, The Psalms, etc] – none from Phoenicia; and the greatest/largest monument was built – the Jerusalem temple was 5 times higher than the Pyramids, indicating an advanced nation. This says more likely the Phoenician writings came from the Hebrew.

    3. There is an inherent doctrine to negate the Hebrew, by Rome, Christianity & Islam, to the extent it is not a good career move to contradict the status quo. Scholars & Historians have equally succumbed to such mendacity; books, encyclopedia, media, Popes & schools, even this respected Archaeology site, all claimed as scholarly today, list Judea in the first century as ‘Palestine’ and Jesus as a Palestinian; a historical impossibility because this name was applied to Judea in the second century, 135 CE, a 100 years after Jesus. Equally, the negation of Hebrew is seen in nominating Aramaic as the spoken language of Jews in Judea 30 [The film ‘The Passion’] – this is false! The Hebrew was forbidden by Nero in the year 66 and the first language of all Jews and their writings in Judea pre-66 was in Hebrew.

  • Joseph says

    ‘ the bible is not a “book” at all’
    This was qualified, namely a multi-page continuing narrative, distinct from a one-paragraph or one-page, and in the alphabetical mode. Its pages can be a book of scrolls or any other form. The Hebrew bible is unique by such criteria and the first one of its kind.

    ‘3) it can hardly be said to have appeared “suddenly”’

    There is no known written Hebrew before the Hebrew bible, not in Canaan or Egypt. Nor is there any older Hebrew seen elsewhere before the Hebrew bible [e.g. in Babylon, etc]; nor was this language spoken or written by any other peoples. its narratives say the Israelites entered Canaan with the five books already completed. There is also the factor of a host of items appearing suddenly with no previous imprints: Monotheism; Creationism; New laws; authentic genealogies of 3,000 previous years; geographical listing of many nations not known of from other sources [the first mention of Philistines, Medianites, etc], the introduction of ‘species’ [emergence life forms in their correct protocol]; that the universe is ‘finite’ [with a beginning]; that light is a primordial product; the epitome of grammar that transcends Shakespeare; still the most impacting book of humanity. It does not get any more ‘suddenly’ than this, even as it comes from a small, desert wandering people who came late in the ancient world: unless a previous work of such magnitude can be exemplified from elsewhere?

    ‘and 4) “older” does not necessarily mean “better.”
    True. However, In this case it clearly does.

  • 1 2

    Some HTML is OK

    or, reply to this post via trackback.

Send this to a friend

Hello! You friend thought you might be interested in reading this post from https://www.biblicalarchaeology.org:
What’s the Oldest Hebrew Inscription? A Reply to Christopher Rollston!
Here is the link: https://www.biblicalarchaeology.org/daily/biblical-artifacts/inscriptions/what%e2%80%99s-the-oldest-hebrew-inscription/
Enter Your Log In Credentials...

Change Password