# The ancient Hebrew letter hiding on the back of a potsherd A pottery fragment from a desert fortress carried a Hebrew message about supplies. New photographs revealed more writing on its apparently blank back. Canonical URL: https://paleopro.app/blog/arad-ostracon-hidden-writing By Eitan Shalev. Published by Paleo Pro Editorial. Published: 2026-10-06 Updated: 2026-10-13 ## A message on a piece of pottery Arad 16 is a broken piece of pottery with a Hebrew message written on it in ink. Archaeologists call an inscribed fragment like this an *ostracon*. Excavators found it in 1965 at Arad, a desert fortress in the ancient kingdom of Judah, in what is now southern Israel. It dates to about 600 BCE, roughly 2,600 years ago. [[1]](https://paleopro.app/blog/arad-ostracon-hidden-writing#source-1) [](https://paleopro.app/assets/blog/arad-16-reverse.png)The back of Arad 16: ordinary photograph (A), infrared image (B), and proposed reading (C). Outlined letters in C are conjectural. [Faigenbaum-Golovin et al., 2017, figure 4](https://doi.org/10.1371/journal.pone.0178400.g004), [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). For decades, readers worked on the writing on the front while the back looked blank. Then Shira Faigenbaum-Golovin and her colleagues photographed the fragment using different wavelengths of light. In their 2017 study, they reported three lines on the back, including a request for wine, along with additional writing on the front. [[1]](https://paleopro.app/blog/arad-ostracon-hidden-writing#source-1) The comparison above is what caught my attention. In the ordinary photograph, I can understand why someone would move on. The infrared view gives the eye something else to follow. That small change opens up a much larger question: how much writing have we missed on objects already sitting in collections? ## What a different photograph can show The team photographed the sherd at several wavelengths. The clearest views of the two sides came from different bands: 830 nanometres for the front and 890 for the back. Their published figures set ordinary photographs beside the enhanced images and drawings of the proposed readings. [[1]](https://paleopro.app/blog/arad-ostracon-hidden-writing#source-1) Keep those three kinds of picture separate as you read. A photograph records a surface under particular conditions. An enhanced image makes some features easier to see. A drawing tells you how the researchers interpret those features. Moving from one to the next is useful, but the drawing should not replace the photograph in your mind. Open figure 4 in the paper and look at the images before looking at the translation. You do not need to read ancient Hebrew to notice the difference between a faint mark, a clear stroke, and a completed letter in a drawing. That is a better first exercise than trying to guess the whole message. ## A supply note is still a piece of literature The later overview by Shira Faigenbaum-Golovin and colleagues places the Arad texts within Judah's military administration. Its comparison with the Samaria ostraca is especially useful: similar-looking groups of short inscriptions can reflect different writing arrangements. At Samaria, the authors identify two scribes across a period of at least seven years; their account of Arad describes writing extending down the military supply system. [[2]](https://paleopro.app/blog/arad-ostracon-hidden-writing#source-2) This gives a supply note a different kind of interest from a royal monument. Instead of asking only which ruler a text names, ask who needed a written message, who received it, and what had to happen after it arrived. Those questions keep the object connected to an activity rather than treating its letters as a decorative alphabet sample. ## How many people could write? A 2020 study by Arie Shaus and colleagues compared forensic document examination with two computer methods. The forensic examination supported at least twelve writers among eighteen examined texts. The two algorithms returned more conservative minimum counts. These are findings about distinguishable handwriting in the studied material, not a census of the population. [[3]](https://paleopro.app/blog/arad-ostracon-hidden-writing#source-3) That difference matters. Twelve hands in a set of letters is not the same as twelve writers living at the place where the letters were found. A letter can arrive from elsewhere. Nor does a collection of military correspondence sample every household, occupation, age, or social group. The authors do discuss wider literacy and the possible setting for composing biblical texts. Those are historical arguments built on the handwriting results. A careful account should preserve that extra step instead of turning a study of handwriting into a date stamp for a biblical book. [[3]](https://paleopro.app/blog/arad-ostracon-hidden-writing#source-3) ## Read the evidence in this order - **Start with the object.** Note its identifier, where it was found, and the date assigned to it. - **Compare the pictures.** Keep the ordinary photograph, enhanced view, and drawing open together. - **Check uncertain letters.** Read the figure caption before assuming that every drawn stroke survives on the pottery. - **Read the argument last.** Separate what became visible from what the researchers conclude about the people who wrote it. If the letter shapes are unfamiliar, keep our [Paleo-Hebrew alphabet chart](https://paleopro.app/alphabet) beside the paper. Use it to learn the letter correspondences. A modern font gives you tidy examples; an ancient hand gives you an individual act of writing. The reward in this paper is small and very human: a message that had more to say than anyone could see. Its discovery also gives the reader a practical habit. Before accepting that a surface is empty, ask how it was examined. ## Arad 16: the request for wine Opening words on the reverse, following the 2017 proposed reading, figure 4C. The photograph above shows the rest of the fragment and its uncertain letters. ```text אמ יש יינ שלח ``` ## Sources 1. [Shira Faigenbaum-Golovin et al. (2017). Multispectral imaging reveals biblical-period inscription unnoticed for half a century. PLOS ONE 12(6), e0178400.](https://doi.org/10.1371/journal.pone.0178400) Original imaging study. Figures 3 and 4 distinguish photographs from proposed readings. 2. [Shira Faigenbaum-Golovin et al. (2021). Literacy in Judah and Israel: Algorithmic and Forensic Examination of the Arad and Samaria Ostraca. Near Eastern Archaeology 84(2), 148-158.](https://doi.org/10.1086/714070) Comparative overview. The publisher provides an abstract; full text may require access. 3. [Arie Shaus et al. (2020). Forensic document examination and algorithmic handwriting analysis of Judahite biblical period inscriptions reveal significant literacy level. PLOS ONE 15(9), e0237962.](https://doi.org/10.1371/journal.pone.0237962) Open-access handwriting study. See the results and discussion for the counts and their interpretation.