Senzar is the name given to a sacred language that is now lost from public view and has become secret. Our only source on this language is the writings of H. P. Blavatsky. According to her, Senzar is the language in which the “Book of Dzyan” was recorded, from which she translated the stanzas that form the basis of her book, The Secret Doctrine. It is there described as a pictorial language of symbols, and this is how it has come to be thought of among students of Theosophy. However, in some places she also described Senzar as a phonetic language. With the publication in 2010 of The Secret Doctrine Commentaries that Blavatsky had given in 1889, but that had remained unknown for 120 years, no doubt could any longer remain. The stanzas she translated were from the phonetic form of Senzar, not the pictorial form. The idea that Senzar is solely a pictorial symbol language has hindered research on it for all these years. Once we begin looking for its phonetic form, we find clear evidence for the existence of this lost sacred language.
The “Archaic Manuscript” written in symbols that Blavatsky vividly describes at the beginning of the The Secret Doctrine (vol. 1, pp. 1-5) is not the “Book of Dzyan” that she translated stanzas from. She makes this clear in another place, referring to the “one small archaic folio” as “the Book of the Secret Wisdom of the World,” and describing the “Book of Dzyan” as the first of fourteen volumes of commentaries on it (Blavatsky, Collected Writings, vol. 14, p. 422). That these fourteen volumes of commentaries are written in a phonetic form of Senzar rather than in pictorial symbols could be deduced from Blavatsky’s statements made in 1888 in The Secret Doctrine, and this was confirmed in The Secret Doctrine Commentaries published in 2010. She refers several times to the words of the “Book of Dzyan,” phonetic words; says that she has tried to give a verbatim or word for word translation; and refers several times to specific numbers of verses in the original “Book of Dzyan” that she has omitted. These, of course, would be phonetic verses, consisting of phonetic words, not pictorial symbols.
According to The Secret Doctrine (vol. 2, p. 200), the first language of the fifth root-race was the inflectional speech, and this is “now the mystery tongue of the Initiates,” i.e., Senzar. It is there described as “the root of the Sanskrit, very erroneously called ‘the elder sister’ of the Greek, instead of its mother.” In her 1877 book, Isis Unveiled (vol. 1, p. 440), Blavatsky had described Senzar as “ancient Sanskrit.” It was described by “a Chela” in 1883 as “the direct progenitor of the Vedic Sanskrit,” adding that “the sacerdotal speech of the initiated Brahmin, became in time the mystery language of the inner temple, studied by the Initiates of Egypt and Chaldea; of the Phoenicians and the Etruscans; of the Pelasgi and Palanquans, in short, of the whole globe” (“Was Writing Known before Panini?,” The Theosophist, vol. 5, 1883, p. 18, reprinted in Blavatsky, Collected Writings, vol. 5, p. 298). We thus learn that Senzar was an inflectional language, described as “ancient Sanskrit,” as “the root of the Sanskrit,” and as “the direct progenitor of the Vedic Sanskrit,” and that it was once in use as a sacred language across the whole globe.
The linguistic term “inflectional” describes languages whose words undergo change in order to give grammatical information, usually by way of inflectional endings (verb conjugations and noun declensions). These inflectional endings characterize the languages that comprise what is today known as the Indo-European language family. This family includes the ancient languages Sanskrit, Avesta, Greek, Latin, etc., and the modern languages that descended from them, Hindi, French, German, English, etc. The ancient Indo-European languages are thought to have all descended from a hypothetical Proto-Indo-European in the even more ancient past. This matches the teaching of The Secret Doctrine that the inflectional speech was the first language of the fifth root-race. Phonetic Senzar, then, would be a sacred form of what is today called Proto-Indo-European. While there is much evidence for the existence of Proto-Indo-European, is there any evidence for a sacred form of it?
Of course, the most well-attested and well-preserved ancient Indo-European language is Vedic Sanskrit, which is indeed a sacred language. Its sister language Avesta is also well-attested, again by way of a body of sacred writings. But is there any remnant or trace of a language that would be “the direct progenitor of the Vedic Sanskrit,” as Senzar is said to be? According to Indian tradition, the Vedas were seen or heard by ancient seers, and have been preserved unchanged since then. They did not develop from anything. They had no progenitor. Such is the traditional view. An unexpected fact, however, has long been noticed. We find that many Vedic verses are repeated in the various Vedic texts, and sometimes they show variations that cannot be attributed to scribal error. Maurice Bloomfield in his 1906 Vedic Concordance presented a complete “index to every line of every stanza of the [then] published Vedic literature.” Of its about 90,000 entries, about one-third occur more than once. Of these roughly 30,000 repeated verse lines, about one-third show variants. So of about 90,000 verse lines, about 10,000 show variants. These were studied in three volumes of Vedic Variants, 1930-1934. One in nine is a lot of variants, far more than would be expected if the Vedic verses in fact had no predecessor or progenitor.
Even more disturbing to the traditional view is the finding of Prakritisms in the Ṛgveda. The Ṛgveda is the oldest, most sacred, and most perfectly preserved of the Vedas. Its language should consist entirely of sacred Sanskrit; there should be no trace of any vernacular Prakrit in it. Yet this is what modern research is finding (e.g., “Prakritism in the Ṛgveda,” by G. V. Devasthali, 1970; “About the Traces of a Prakrit Dialectal Basis in the Language of the Ṛgveda,” by T. Y. Elizarenkova, 1989; “Prakritic Wordforms in the Ṛgvedasaṃhitā,” by Chlodwig H. Werba, 1992). What does this mean? First we saw clear evidence that the Vedas have predecessors or progenitors (as would be assumed by the modern linguistic theory of a Proto-Indo-European). There must have once been many more Vedic texts than are now preserved. Then, at least some of these would have been in a Prakrit or included Prakrit words, phrases, and idioms. If the Prakrit languages developed only later than Vedic Sanskrit, as has been generally assumed, it would be hard to explain the presence of Prakritisms in the Ṛgveda. The Ṛgveda has been preserved with such scrupulous accuracy that these Prakritisms are unlikely to be later modifications introduced into it, but rather were there all along.
The Prakrits are mostly thought of as vernacular or everyday languages, in contradistinction to sacred languages. There are, however, two major exceptions. The sacred canon of the Śvetāmbara Jainas is written in the Ardha-Māgadhī variety of Prakrit, and the sacred canon of the Theravāda Buddhists is written in Pali, which can linguistically be considered a variety of Prakrit. The general idea is that these are vernaculars that came to be thought of as sacred languages because the sacred books of these two traditions have come down to us in these languages. Buddhists say that the Buddha Gautama purposely taught in the vernacular language of his time and place, rather than in the sacred Sanskrit language, so that the people could understand him (e.g., Cullavagga 5.33). Śvetāmbara Jainas say that the Jina Mahāvīra taught in Ardha-Māgadhī, which was the vernacular of his time and place. It was, however, understood by the various hearers in their own language (Aupapātika-sūtra 56). Digambara Jainas say that the Jina taught using the “divine sound” (divya-dhvani), and that his gaṇadharas, his close disciples who could understand this, translated it into the vernacular of their time and place. Thus we have sacred canons written in vernacular languages that became sacred languages. In both traditions, however, there is an alternate view.
Some Buddhist and Jaina writers held that Prakrit is the original language, and that Sanskrit came from it, not vice versa. The information and sources on this were summarized in a 1993 article by Johannes Bronkhorst. He writes: “Māgadhī, we read in Buddhaghosa’s Visuddhimagga, is the original language (mūlabhāsā) of all living beings, . . .” (p. 398). Māgadhī is held to be Pali by the Buddhist commentators, the language of the canon, and this is a Prakritic language. A later Buddhist writer says that “all other languages are derived from Māgadhī,” including Sanskrit (p. 399). Some Jaina writers have likewise held that their sacred language, the Ardha-Māgadhī variety of Prakrit, is the original language, and that Sanskrit comes from it (pp. 399-401). Even the Hindu writer Bhartṛhari, after noting in his Vākyapadīya that the divine language Sanskrit has been corrupted by incompetent speakers, tells us that the upholders of impermanence (apparently Buddhists) say the opposite (p. 406). That is, according to the ancient vṛtti thereon, Prakrit is the correct language and it has been altered to become Sanskrit (p. 407). Bronkhorst’s article is suggestively titled “Buddhist Hybrid Sanskrit – the original language,” although he did not actually make this claim. We now continue with this intriguing topic.
Buddhist Hybrid Sanskrit is a name coined by Franklin Edgerton to describe the language of the Sanskrit Buddhist texts. These Sanskrit texts, which include the Mahāyāna Buddhist texts, form a Buddhist canon distinct from the Pali Buddhist canon. In a 1936 article, “The Prakrit Underlying Buddhistic Hybrid Sanskrit,” Edgerton postulated a “protocanonical Prakrit” on which Buddhist hybrid Sanskrit was based. He analyzed the language of the Sanskrit Buddhist texts to determine which features distinguish it from Classical Sanskrit. He spent the rest of his life studying and describing these distinguishing features, culminating in his monumental Buddhist Hybrid Sanskrit Grammar and Dictionary, 2 volumes, 1953. These Prakrit features that distinguish Buddhist Sanskrit from Classical Sanskrit also distinguish it from any other specific Prakrit known, including Pali (1936, pp. 509, 516). Therefore he had to postulate an earlier Prakritic language that both Buddhist Hybrid Sanskrit and Pali were based on. There must have existed a considerable body of canonical or sacred texts in this Prakrit (p. 502). He called this Prakrit “protocanonical”; “proto” in that it is a hypothetical language, and “canonical” specifying its use in sacred texts. So here we have a sacred form of Proto-Indo-European, perhaps the very one we were looking for.
The Prakritisms found in Vedic Sanskrit now take on a new significance for us. The possible relationship between these Prakritisms and the protocanonical Prakrit behind Buddhist Hybrid Sanskrit has not yet been explored, no doubt because the Vedas are considered much older than the time of the Buddha. Theosophy, however, accepts the traditions of previous Buddhas, and therefore of a previous canon of Buddhist texts. These Prakritisms aside, Sukumar Sen noticed long ago an important fact regarding similarities between the syntax of Buddhist Sanskrit and Vedic Sanskrit, which he calls Old Indo-Aryan. In his 1928 article, “An Outline Syntax of Buddhistic Sanskrit: Being a Contribution to the Historical Syntax of Indo-Aryan,” he writes (pp. 1-2): “The third division is the Buddhistic Sanskrit properly called. It is generally known as the ‘Gāthā language,’ or as ‘Mixed Sanskrit.’ Its philological importance is of the utmost. From the syntactical point it is doubly interesting, as it retains much of the remnant of Old Indo-Aryan idioms which were lost in the classical Sanskrit, . . .” Edgerton did not deal with syntax in his Buddhist Hybrid Sanskrit Grammar, and this important observation of Sen’s also remains to be fully explored (Elizarenkova devotes pp. 14-16 of her above-mentioned article to syntax).
The possible relationship between a progenitor of Vedic Sanskrit and the protocanonical Prakrit behind Buddhist Sanskrit is not the only evidence we have among the Hindu Sanskrit texts. Before Buddhist Hybrid Sanskrit was labeled as such and studied, F. E. Pargiter had made a very detailed study of The Purāna Text of the Dynasties of the Kali Age (1913). He found clear evidence that in the oldest Purāṇas the verses had been Sanskritized from an earlier Prakrit. This is exactly what Edgerton later found in the Sanskrit Buddhist texts. Further, Pargiter described this Prakrit as “a literary language not far removed from Sanskrit” (p. xi). Similarly, Edgerton found that the Sanskrit Buddhist texts were not just translations or re-workings of Pali originals (as some writers had supposed, p. 502), because the Sanskrit elements in them were as original as the Prakrit elements (pp. 508-509). The protocanonical Prakrit behind the Buddhist Hybrid Sanskrit texts and the old literary Prakrit behind the oldest Hindu Purāṇas, the one we are looking for, would not be a Prakrit that descended from Sanskrit, i.e., not a Middle Indo-Aryan language that descended from Old Indo-Aryan as are the Prakrits now known. It would be an earlier proto-Sanskrit that had some of the features now found or retained only in the Prakrits, features that were removed from this proto-Sanskrit when it became Sanskrit, “refined,” “polished,” “perfected.”
We have now seen clear evidence for the existence of a lost sacred language that we may call Senzar. When Senzar is regarded solely as a pictorial symbol language, there is not much to find. When we look for a phonetic form of Senzar that is a precursor to Sanskrit, an inflectional language described as “ancient Sanskrit,” as “the root of the Sanskrit,” and as “the direct progenitor of the Vedic Sanskrit,” there is much to find. We then find that what is apparently just such a language has left major traces in the Buddhist Sanskrit texts. Its features that differ from Classical Sanskrit have been analyzed by Franklin Edgerton and described at length in his Buddhist Hybrid Sanskrit Grammar and Dictionary. To explain the observed evidence, Edgerton postulated a protocanonical Prakrit predecessor of both the Sanskrit and Pali of the Buddhist canons, that would also be close to the Ardha-Māgadhī of the Jaina canon. Such a language has left similar traces in the oldest Hindu Purāṇas. It has apparently even left traces in the ancient and sacrosanct Vedic texts. It would be a sacred form of Proto-Indo-European, just what we would expect a phonetic form of Senzar to be.