The closing words of the Old Testament are a prophecy about the Lord sending Elijah, who “shall turn the heart of the fathers to the children, and the heart of the children to their fathers, lest I come and smite the earth with a curse” (Malachi 4:6). That verse is important to members of The Church of Jesus Christ of Latter-day Saints, for we believe that Elijah literally was sent as an angel to Joseph Smith in a dramatic event in which priesthood keys related to that mission were given to Joseph Smith. Our significant efforts related to family history research and the work done in temples all over the earth are tied to that sacred  event. Now if you want to turn your heart to your ancient fathers and mothers, one way to do that is to learn about them from their own words, when available. But what happens when the only words we have from them are written and preserved in a script that nobody has been able to decipher? That’s the situation for the ancestors of what may be a large fraction of the earth, including many at least in Asia, Europe, and the Americas, who share ancient connections with the one of the ancient world’s greatest but most mysterious civilizations, the Indus Valley Civilization.

The Indus Valley Civilization flourished from about 2600 BC to 1900 BC, but was in existence around 3300 BC and persisted until roughly 1300 BC.  It was spread across what is now northwest India, central-eastern Pakistan, and part of Afghanistan, centered around the lower regions of the great Indus River. The civilization is also called the Harappan Civilization after the archaeological site in modern Punjab, Pakistan where research the civilization began in the 20th century. Archaeologists have found that the Indus Valley Civilization was impressive in many ways, with advanced drainage systems, urban planning, metallurgy, etc. While once large and mighty, it strangely disappeared and has remained invisible to us until very recently. At its height, though, it could be compared to ancient Egypt and Mesopotamia, a great civilization that may have influenced many other lands and many later peoples, especially in India. Their descendants may include many of us. There’s certainly been time for their genes to spread to many lands.   

They had a written language preserved in hundreds of short messages, often on soapstone (steatite) seals, typically just a few characters long. A few samples are shown below.

Surely there are more writings to be found, perhaps including some treasures of much more extensive work. Or maybe what we have is pretty much all that those who could write put down on permanent media. In any case, knowledge about the millions who were part of the Indus Valley civilization over the centuries has been hindered by the mystery of their script which has remained undeciphered. Undeciphered until now, perhaps, thanks to a remarkable breakthrough by a linguist, Steven C. Bonta, who happens to be a member of The Church of Jesus Christ of Latter-day Saints, and feels  a sense of duty to help the world remember and perhaps even turn their hearts to the millions from that great civilization. He has been working on this project for over 30 years ever since he made initial efforts to understand the script as a young graduate student. Steve worked on the Indus Valley script for his master’s degree at Brigham Young University. After receiving his Ph.D. in linguistics at Cornell, where he worked on a Tamil dialect, he has continued puzzling over the mysterious script over the years, but had an especially productive era during his recent years in China, which is where I had several opportunities to talk with him. COVID resulted in more time to focus and explore new angles with the Indus Valley script.

A few weeks ago, Steve felt that his work was far enough along that it was time to begin sharing it with what he expected would be a skeptical world, and I was privileged to listen to him given a lengthy PowerPoint presentation in our home. It was the most exciting presentation I can recall sitting through, in spite of being the longest — three hours, though one hour of that was probably due to my numerous questions. I am not a linguist, but have been fascinated by the few languages I’ve studied over the years and the mysteries behind them and their writing systems. I’ve also seen questionable efforts to crack undeciphered scripts, including various efforts to find parallels in various scripts to the characters copied from the golden plates of the Book of Mormon and the efforts to understand Egyptian reflected in the Kirtland-era Grammar and Alphabet of the Egyptian Language that seems to have been compiled after Joseph Smith’s revealed Book of Abraham was given in an effort to link that revelation to the Joseph Smith Papyri and learn how Egyptian worked. Steve’s methodology was radically different and much more satisfying and meticulous.

By painstaking analysis of the patterns that the characters make, he was able to identify several different structures that can help point to the role the characters are playing, such as serving a notational function (like our “$5” which means “five dollars” — written in the opposite order to the spoken language) or a transcriptive function that conveys language as spoken. His approach involves the following:

• Ascertaining the nature of the writing system
• Ascertaining whatever can be learned about the typology of the underlying language
• Becoming completely familiar with the signs and their patterns of distribution
• Exploiting the rhebus principle in instances where both the distribution and the graphology of a sign happen to suggest a particular value
• Making use of any contextual clues
• Trial and error: Laboriously trying values for signs with sufficient available distributional, contextual, and/or graphological data

A key feature when language breakthroughs are made, such as finding a previously unrecognized connection between two languages, tends to be bursts of explanatory power in which solving one puzzle gives new insights that then can be used to solve other mysteries and create coherence where mystery once reigned. During Steve’s presentation, there were multiple “ah hah!” moments with these bursts of explanatory power, adding the sense that these new advances are converging and giving a firm new framework for the ongoing work that will be needed.

In general, I found his work to be cautious, meticulous, logical, and well supported. He recognizes the gaps within the work and the need for much more research, but being able to provide plausible, self-consistent analysis that provides good explanatory power with multiple layers of confirmation is an impressive feat, especially in light of the barriers that had to be overcome. The decipherment is partial —  60 or so characters have been deciphered, but it is enough to find many names and titles of the owners of many of the inscriptions. Some just have the name of the owner, and others also add add notational items that appear to describe what is being donated to a temple. This opens up doors to understand the caste system of the Indus Valley people and the role that prominent individuals played, and perhaps the importance of temple worship.

The work of cracking ancient scripts tends to be the territory of professional archaeologists, so there is always skepticism, often appropriately so, when an outsider claims to have a breakthrough. I hope that Steve Bonta’s work will get a fair hearing and not be simply dismissed for years as was the case when a Russian outsider, Ukrainian-born Yuri Knorozov, found tantalizing evidence that Mayan glyphs were not necessarily pictographs or rebuses that depicted concepts, but could represent phonetic units. The dominating archaeologist ruling the world of  Maya studies, J. Eric S. Thompson of England, denounced his work, and few others in the academic world dared resist.  As Harri Kettunen, president of the European Association of Mayanists, put it, “[Thompson] dominated and nobody objected to him. It was like with Marx: You couldn’t oppose.” See Coilin O’Connor, “Yuri Knorozov: The Maverick Scholar Who Cracked The Maya Code,” Radio Free Europe/Radio Liberty (rferl.org), Nov. 20, 2022.

The decipherment of Mayan was greatly aided by the existence of the modern Maya language in spoken form that was obviously related to that of their ancestors a few centuries ago. It was also aided by a Spanish priest, Diego de Landa, who had recorded information about the language obtained from someone who knew the written language. We have no such luxury with the Indus Valley Script. We also have nothing like the Rosetta Stone, that was a key to cracking the Egyptian code. And unlike Egyptian or Mayan,  the Indus Valley script so far lacks any lengthy writings to help us analyze the language.

One of the biggest barriers to decipherment, in my opinion, is that we don’t even know which language family is most closely related to the Indus Valley script. There have been somewhat heated debates over this basic aspect of the Indus Valley language. Was it a precursor of the Indo-Aryan language family that includes many languages found in northern India and its neighbors such as Hindi, Bengali, Gujarati, Punjabi, and Urdu? Or was it part of the Dravidian language family, spoken primarily in southern India and southwestern Pakistan, including Telugu, Tamil, Kannada and Malayalam? Regional politics could make any answer a sensitive issue, but academic politics may be a bigger issues, based on what I’ve seen in the academic world from time to time.

An excellent article on the Indus Valley script in the prestigious journal Nature offers the dominant academic perspective today, especially in the West: “As for the language, the balance of evidence favours a proto-Dravidian language, not Sanskrit. Many scholars have proposed plausible Dravidian meanings for a few groups of characters based on Old Tamil, although none of these ‘translations’ has gained universal acceptance.” See Andrew Robinson, “Ancient civilization: Cracking the Indus script,” Nature 526 (2015): 499–501, https://doi.org/10.1038/526499a. The academy favors a connection to Dravidian, while Steve Bonta’s work provides compelling evidence for a connection to Sanskrit, an Indo-Aryan language. I hope his work will receive proper attention in spite of the sensitivities it may stir.

Steve has prepared a short paper on his work for the traditional peer-review process. Meanwhile, to make the details of his work available in a way that can’t fit in the short space that journals provide for a paper, a much longer paper (259 pages) has just been posted at Academia.edu. See Steven C. Bonta, “A Partial Decipherment of the Indus Valley Script: Proposed Phonetic and Logographic Values for Selected Indus Signs and Readings of Indus Texts,” Academia.edu, August 1, 2023, which can also be accessed at the preceding link or via a shortcut I prepared: https://tinyurl.com/indus-script. There is also an Aug. 1, 2023 letter that he sent to a number of people explaining his work and summarizing his methods and major findings in few pages. You can read the introductory letter online. If Steve is right, and I think there is a high probability that he is given the multiple lines of confirmation for the direction of his work, this may be a monumentally important advance in deciphering a lost language and opening doors of discovery for a once-lost civilization. Please share this information with anybody who might be interested. 

If the approach and findings hold up to scrutiny, next steps will include expanding the search to decipher further characters. AI tools may be useful here, for there is a great deal of tedious work needed to test hypotheses and look for meaningful correspondences. But I am extremely excited about what Steve has accomplished. Yes, as a friend and fan, I’m biased, but I’ve been more than happy to tell many friends I admire when I think they are pursuing a fundamentally flawed direction or when their conclusions stretch credulity. I see something much more promising here.

Below are the final two paragraphs of the introductory letter. They address the reasons for publishing a paper prior to completing peer review and also address the potentially sensitive issues around Sanskrit rather than the Dravidian language family as a key connection to the Indus Valley language:

I realize that it is customary to seek publication before going public with potentially important results. Accordingly, a paper much shorter than the attached monograph (though still quite lengthy) is now under submission at a peer-reviewed journal. But being well-aware of the history of decipherment and the very high barriers of skepticism and even ridicule that often must be overcome, I am under no illusions as to the likelihood that the enclosed results, in their full form, will be publishable in any traditional sense anytime soon. This is owing not only to the results, which are bound to be controversial, but also to the fact that it is impossible to make a convincing case for them within the usual 4000-6000 word limits typically imposed by scientific journals. And while there are certainly errors in the enclosed work, I am confident enough that the conclusions are mostly correct that I’ve decided to share them with those most likely to have a genuine, unbiased interest in testing them. I have also posted the monograph to academia.edu, so I consider this now to be public information; feel free to share this email or the attached monograph with any others who might be interested, and feel free to cite with attribution. As far as I’m concerned, these results, correct or not, belong foremost to the people of India and greater South Asia, and in this spirit, I would rather they have unfettered access to them sooner than later.

Finally, nothing in these results is the result of aspirational biases or preferences. As one who did his PhD research at Cornell documenting a dialect of Tamil, who has spent years both studying Dravidian languages and publishing in the field of Dravidian linguistics, and who harbors a deep and abiding love for the magnificent Tamil language and culture, no one would be happier than me to discern a Dravidian solution to the Indus Valley script. But the information available, it seems to me, leads ineluctably to a different result, one which, I hope, I have sufficiently documented, both as to readings and to methodology, to allow for reproducibility (or falsification, as the case may be). Although the Indus corpus holds no votive hymns, royal edicts, or literary fragments of the sort that make compelling historiography, it does disclose to us the names and titles of many forgotten personages, and it reveals critical information about their economy and at least one of their languages. I hope that this work may contribute to the progress of Indus research, so that as full a decipherment as possible may be achievable before many more years have passed.

Getting back to the challenges of deciphering other scripts, including the fascinating characters from the gold plates of the Book of Mormon on the “Charactors” document that Martin Harris showed to the learned Charles Anthon, Steve Bonta’s presentation included a slide about decipherment methods that don’t work:

Decipherment: What DOESN’T Work

Two time-dishonored methods of decipherment:

1.Trying to identify the graphology of the sign and assigning it the meaning of whatever the sign “looks like.”

2.Having already decided what the target language must be, trying to identify the graphology of the sign and assigning it the sound of whatever the sign is assumed to represent in the target language (the WRONG way to exploit the rhebus principle).

Thinking of the small sample of Book of Mormon characters that were copied from the gold plates, I asked Steve if he thought his methodology, perhaps coupled with AI tools, could lead to some improved insights. He didn’t think so. We likely do not have enough text to work with. But I hope that perhaps there will be paths forward in the future, especially if additional characters become available.

Speaking of writing on metal plates, what Wikipedia claims is the longest known Indus Valley inscription in their article “Indus Script” has just 34 characters engraved on a copper plates, which sounded quite interesting to me when I first read that. However, it has unknown provenance and is likely a fake, as Steve pointed out to me after I included Wikipedia’s photo of that questionable artifact in the original version of this post earlier today (sorry!). Exotic artifacts from the Indus Valley civilization apparently are being cranked out by craftsmen at a rapid rate to exploit gullible collectors. Writing on copper plates did occur in some parts of India anciently, but not that we know of in the Indus Valley civilization.

Many thanks to a persistent, meticulous, and brilliant linguist for the years of effort that appear to be bearing fascinating fruit about an ancient civilization that, like some others we may know of, seemed to disappear from sight for many centuries. They may have been the ancestors of many of us. I look forward to learning more, and hope that more extensive documents will yet be found and understood.