Tanakh contains more than a few hapax legomena (words that appear only one time in a given literature). This poses a potential problem for us, because when a word appears only one time it is hard to know exactly what it means. In theory, we should have a masorah for each word in Tanakh, but this isn't always the case, although Rav Saadya showed the value of masorah in his work Tafsir al-Sab'ina Laf?ah min Mufradat al-Quran (פתרון שבעים מילים in Hebrew). Encylopedia Judaica quotes an excerpt:
"They [that is, the Karaites] are unaware of the fact that they have come to know the sense of these words only from what I have adduced as proof and thus revealed their meaning from the Mishnah."What Rav Saadya wrote is true, but not every lonely word in Tanakh is explained in rabbinic literature nor does it seem to have been maintained orally.
Nevertheless, there are options for when this occurs. Clues such as context and the possible etymology of the word may give us the true meaning. When this occurs, the meforshim do precisely this. Many biblical hapaxes were deciphered by the meforshim in this way. Sometimes modern scholarship is able to give us the meaning of a hapax with a high degree of probability. A very neat example where this occurred and it is probably so close to certain as to be simply true is on 1 Sam. xiii 21, which contains one hapax in particular that remained obscure for a long time. Chazal never mentioned it, R. Saadya didn't mention it. Rashi and the Radaq do, as we will see. But archaeology seems to have nailed it.
The background is the oppression the Pelishtim subjected the Benei Yisrael to. The Pelishtim seemed to have controlled the iron smithing market and as a consequence only Sha'ul and Yehonathan had swords and spears! Everyone else had farming tools: axes and the like. Since there were no smiths in Israel, they would go to the Pelishtim to sharpen their farming tools.
Here is the text, with the new JPS translation:
The words וְהָיְתָה הַפְּצִירָה are not difficult to translate, except for what comes afterwards: פִים.
This is the only time the words appears and context doesn't really give it away. Lacking a masorah about the word, Rashi and the Radak after him realized (correctly) that the root of פִים must be פה. Pursuing a mouth-oriented explanation of the word, they concluded that this word was some type of plural for mouth and therefore it must be another sharp tool for sharpening. Rashi writes:
לימ"א בלע"ז שיש לה הפצר פיות פיות וחידודיה הרבה היתה להם לחדד המחרשת והאתים
Some early English Bible sort of glossed over the problem. For example, the 1384 Wycliffe Bible translated "Thanne al Yrael descendide to Philistiym, that echon sharpe his shaar, and diggyng yren, and axe, and purgyng hook." Or one from 1535 (Coverdale): "And all Israel were fayne to go downe to the Philistynes, whan eny man had a plowshare, a mattock, an axe, or a sythe to sharpe: and the edges of the plowshares, and mattockes, & forckes, and axes, were laboured, and the poyntes blont." (Okay, I added all that mostly because its fun.)
In any case, by the time the early 17th century rolled around, English translators of the Bible were looking at Rashi and the Radaq and the King James Version produced this: "Yet they had a file for the mattocks, and for the coulters, and for the forks, and for the axes, and to sharpen the goads."
In other words, וְהָיְתָה הַפְּצִירָה פִים was rendered as Yet they had a file in English (I think the earlier Tyndale also translated this way, but I wasn't able to check).
This translation sufficed for a very long time, and in some circles still suffices, although it was recognized that the meaning of פִים still could not have been said to be certain, there were grammatical issues with its derivation.
In 1904 George A. Barton published a short article called "Two New Hebrew Weights" in Journal of the American Oriental Society, Vol. 24. (1903), pp. 384-387. It began and looked like this:
The article continued to discuss what the three letters visible on the left say and what they mean. He readily identifies two of the letters as י and מ but has a little trouble with the third (actually, first) letter. He thinks it might be a פ but that it also mean be a ג given its similarity with some Hasmonean era examples of paleo-Hebrew גs. He thinks it may be an abbreviation, and speculates about that. Maybe it stands for גרה י למלך ten royal gerahs. He then explores whether it is a פ, but unfortunately still assumes that its an abbreviation. A lot of great pilpul can come from a faulty original assumption, so Barton's surmise that it may stand for לפי משקל is tasty, but it apparently didn't occur to him to dwell on the weight of the weight or on whether there isn't some instance of פים somewhere that might be of assistance. Even worse, he ends by assuming that it being a ג is more likely since he would rather know that this is a gerah weight.
Ten years later E. J. Pilcher published an article in Palestine Eploration Fund Quarterly Statement called "A New Hebrew Weight," on the very same matter. Pilcher quotes someone called Samuel Raffaeli who owned such a weight and suggested that this weight was a!פים!
The weights were since weighed, compared, contrasted, speculated and thought over until it was realized that this weight was in fact equal to 2/3rd of a sheqel. You see, there were weights weighing (and marked) sheqel found to compare it to. In ancient Israel, when you wanted to pay someone a sheqel you could take gold and a one sheqel weight and put them on scales. When you wanted to pay a half sheqel you could do the same. And when you wanted to pay 2/3rd of a sheqel, you wanted to pay a pim.
And that, it seems, is exactly what verse 21 of 1 Sam. xiii is saying. The Benei Yisrael had to bring their tools to Pelishtim for sharpening and they had to pay a fee which amounted to one pim. In light of this, some new translations of Tanakh say something like "and the price of the filing was a pim for the mattocks," which is what the new version of the JPS does, neatly incorporating the old (filing) with the new (price...was a pim).
What I will not got into here is whether or not the original biblical vocalization of פים was payyim, given that in biblical times matres lectionis were rarely or scarcely used, which would mean that a biblical-era artifact marked פים would probably be spelled פיים today, since if it had been pronounced פים it would probably have been spelled פם. Nearly all examples of biblical-era Hebrew are written haser, so the י here just might be a consonant rather than a vowel.
It would seem almost certain that we now know what a pim is. In this post I discussed how Ramban used what we would today consider an archaeological finding in aiding his own understanding of Torah (in this case, how much a sheqel weighed).
An interesting outgrowth of this finding is that we now know that the root פה relates to thirds and this might give us a new understanding of the biblical expression פי שנים, the double portion due firstborn sons. For more, an interesting lecture by R. Yitz Etshalom can be heard on the OU's website.
Should new translations use this? Following the example of the Ramban, I believe they should--or at least they should use be as willing to use this finding (which is sort of old news by now) as much as the exact language of the King James Version, even though it is based on Rashi and Radaq. A Tanakh by a publishing house which shall not be named yet naturally eschews any reference to this topic and translates ala Rashi/ Radaq via KJV. But a post about it will be forthcoming. Note in advance: it is very, very difficult to fault a translator for using Rashi. Nevertheless, one must remember that Rashi used his best abilities to arrive at an enduring translation of a very difficult word. It is his example that we need to continue to follow.
(It goes without saying that this is a brief discussion of the subject. A lot more could be said on the subject, both the substance of the post, including implications for other pesukkim and the last paragraph. For a radically different view, see here)