top of page

The (Ivory) Tower of Bavel

Examining the Torah’s account of language development from the perspective of modern Linguistics.

Emmett L. Stone

Contrary to popular belief, the scientific understanding of language development is not at odds with the Torah’s account of the creation of language or its initial dispersal. We shouldn’t expect it to be, of course. As with any social science, linguistics is often narrative in the way we comprehend it, and this will overlap with history and culture.

Take for example that linguistics uses comparative methods — looking at grammar, morphology, historical data, and sometimes basic vocabulary — to reconstruct what languages looked and sounded like in past eras. More than that, all linguists see that the modern set of languages we have are descended from other specific parent languages. In this way, English, Latin, Greek, Persian, Sanskrit, and many more all had one common ancestor, now called Proto-Indo-European. Likewise, Mandarin and Tibetan had another, or Finnish and Hungarian with yet another common ancestor, and so on around the globe.

All of these languages, some now reconstructed with linguists’ best guesses for proto-languages, are traced back to roughly the same point, five-thousand some-odd years ago. The exception for this five-thousand some-odd years mark is found with the Semitic languages, where Hebrew and Aramaic are classified, traced back almost double as far. All of these proto-languages now have many descendants that are each unlike the rest. It is not a great leap to say that these relatively few parent-languages existed at the time of the Tower of Bavel and the events described in Sefer Bereshis.

The question of just how these languages came to be, though, is actually more debated among the Rishonim and Achronim - the leading rabbinic scholars since the 11th century AD.

For instance, if the world was created in Hebrew (Bereshis Rabba), did all the other early languages exist at the same time with everyone understanding each of them until they were scattered (Megilla of Talmud Yerushalmi)? Or was there only one language, with the rest spontaneously created after the Tower of Bavel incident (Koheles; Rashi)? Alternatively, were the people just geographically scattered with language diverging naturally (Ibn Ezra)? Perhaps even a more radical view is that even the Biblical Hebrew we know was not the same as what Hashem used to create the world (Di Kasif). We can live comfortably having a set of tenets from the Torah without perfect clarity on the mechanics, as each of these views has its own rationale and sources of evidence. One thing to bear in mind, though, is that Chazal, our sages, certainly used the principles to understand Torah holistically.

In Meseches Sukka (35b), there is a discussion on how to interpret, and indeed pronounce, the Hebrew line פרי עץ הדר (pri etz hodar), ordinarily rendered as ‘fruit of a beautiful tree’. The debate is fascinating and deserving of thorough analysis unto itself, but in it Ben Azzai states it should be read as אִידוֹר (‘idur) instead of הָדָר (hodar) from the Greek word meaning ‘water’, ύδωρ (ydor), or in Ancient Greek ὕδωρ (hydōr). This is by no means the only occasion wherein a verse from the Tanakh is understood by means of other languages. Earlier in the mesechta, the same process is done from an Aramaic source.

This form of drasha (logical step) is only possible with the understanding that all languages descend from Hebrew, at least in some form or another. Hebrew was the language in which the universe was created, the language that Adam HaRishon was given to name what he saw around him (Bereshis), and all other languages developed after this point. Even taking into account that Adam HaRishon spoke Aramaic (maseches Megilla), this still only came later.

This apparent contradiction from Aramaic should rather help to understand that Bavel and its language have always had a special status — the location of Gan Eden; birthplace of Avraham Avinu, the language of the Gemara and some of Tanakh; script of Torah; location of the first exile — and happens to be the most similar language to Hebrew.

It might be tempting to believe that from a modern perspective, these ideas are dismissible. After all, a person today has access to a far greater corpus of information than existed in the past, and perhaps is wary of a notion of Hebrew primacy. That presumption would discredit the wisdom of Chazal and the Torah.

Chazal were certainly familiar with Aramaic, Greek, and the other languages used to make such drushim, and no doubt could see the complete syntactic and lexical dissimilarity between them. That said, it can be a tantalizing endeavor to try and apply all modern science and enlightenment thinking to Judaism. Enlightenment thought, however rational on the surface, is a philosophy that stands on its own legs. One can no sooner prove Torah through Science than prove Buddhism through Islam. Empirical thinking has always existed throughout history and occupies much of the Oral Torah.

While not a perfect analogy, consider the stature of Old English. Despite how much English has seen foreign influence, especially in its vocabulary, everyone would agree that Modern English’s ultimate predecessor is Old English, spoken between the 5th and 11th centuries AD. Nevertheless, no native English speaker can pick up Beowulf in Old English and understand the lines without other training. For instance, it opens with:

Hwæt! We Gar-Dena in gear-dagum, þeod-cyninga, þrym gefrunon, hu ða æþelingas ellen fremedon!

The word order, vocabulary, syntax, morphology, and just about everything else are different to the point of complete unintelligibility between the two Englishes — Old and Modern. There are, in many ways, only traces of Old English carried into Modern English, but the time in-between (approx. between the mid-11th and late 16th centuries) is only a matter of centuries. Like seeing baby pictures of an old man, there is a thread of connection, but the changes are unpredictable and the relation is not always obvious.

So too, we can understand that the languages used five-and-a-half thousand years ago will see wild fluctuations in grammar and vocabulary to the point of vague interrelation, but ultimate nonrecognition. Chazal could come to terms with the fact that examining languages from around the world in a limited way could provide insight - because they knew where to look to find those ancient links. Not only is this concept not at odds with science, but rather it is a rational endeavor that sees the development of all language rooted in one, Divine source.

About the author

Emmett L. Stone

New York born, and British educated, Emmett Stone got a bachelor's in linguistics with German while involving himself in the university Jewish community. He currently lives with his wife in Jerusalem while learning at Yeshivas Ohr Somayach, and runs the linguistics-focused website

bottom of page