"Not created by man"
« previous post | next post »
From Glenn B.:
I just spotted a pair of recently introduced resolutions in the New Jersey legislature that might be of interest to Language Log. SJR 167 and the identical AJR 230 would (if adopted) recognize Sanskrit "as one of the world languages."
Not all of the claims made on behalf of Sanskrit seem kosher to me, particularly the claim that "Sanskrit has a unique origin, not created by man," but I'd love it if Language Log were able to provide a more authoritative discussion.
The resolutions' conclusion
The State of New Jersey recognizes Sanskrit as one of the world languages.
is unobjectionable, but recognition as one of the world's languages is rather weak, given the resolutions long list of odd but strong justifications, and the genuine reasons to celebrate Sanskrit.
The first whereas
WHEREAS, According to the United Nations, Sanskrit has influenced 97 percent of world languages
evokes a United Nations finding that I haven't been able to locate, combining the exact figure of "97 percent" and the vague predicate "influenced".
The second whereas
WHEREAS, Sanskrit is one of the proto-Indo-European languages, similar to Latin and Greek
introduces an inappropriate "proto-", apparently just meaning "old".
The third whereas
WHEREAS, Sanskrit has a unique origin, not created by man, but the language of Rigveda more than 3,500 years ago, making Sanskrit the oldest language of ancient India
is where the "not created by man" parts comes in. That's certainly true, in the sense that Sanskrit was not a conlang, though it has that in common with all other natural languages. The resolution apparently means to imply a divine origin, which seems inconsistent with Sanskrit being "one of the proto-Indo-European languages"…
The fourth whereas
WHEREAS, Sanskrit is the mother of all languages because it is elegant and influential
combines a true statement (that Sanskrit is elegant and influential) with a conclusion (Sanskrit is the mother of all languages) that doesn't seem to follow logically, and is false, if interpreted to mean that all other languages are derived from Sanskrit, and odd, if interpreted in the "mother of all" battles or bombs sense.
I'll leave the rest of the justificatory clauses to readers to discuss, except for noting the sixteenth whereas
WHEREAS, NASA scientist, Rick Briggs, reported Sanskrit is one of the most suitable languages for computers because of its efficiency in developing algorithms
which refers to the article "Knowledge Representation in Sanskrit and Artificial Intelligence", AI Magazine 3/5/1985. The article is genuine, but it doesn't really refer to "efficiency in developing algorithms":
Abstract In the past twenty years, much time, effort, and money has been expended on designing an unambiguous representation of natural language to make them accessible to computer processing, These efforts have centered around creating schemata designed to parallel logical relations with relations expressed by the syntax and semantics of natural languages, which are clearly cumbersome and ambiguous in their function as vehicles for the transmission of logical data. Understandably, there is a widespread belief that natural languages are unsuitable for the transmission of many ideas that artificial languages can render with great precision and mathematical rigor. But this dichotomy, which has served as a premise underlying much work in the areas of linguistics and artificial intelligence, is a false one. There is at least one language, Sanskrit, which for the duration of almost 1000 years was a living spoken language with a considerable literature of its own. Besides works of literary value, there was a long philosophical and grammatical tradition that has continued to exist with undiminished vigor until the present century. Among the accomplishments of the grammarians can be reckoned a method for paraphrasing Sanskrit in a manner that is identical not only in essence but in form with current work in Artificial Intelligence. This article demonstrates that a natural language can serve as an artificial language also, and that much work in AI has been reinventing a wheel millenia old. First, a typical Knowledge Representation Scheme (using Semantic Nets) will be laid out, followed by an outline of the method used by the ancient Indian grammarians to analyze sentences unambiguously. Finally, the clear parallelism between the two will be demonstrated, and the theoretical implications of this equivalence will be given.
Brigg's article is well worth reading, but its ideas about the relationship between Pāṇini's analysis of Sanskit and Semantic Networks (along with other aspects of 1980s AI) applies equally to a Pāṇinian analysis of any language.
JPL said,
July 25, 2025 @ 8:37 pm
So why did the NJ legislature feel the need to "recognize" Sanskrit in this way (or why did the senator and assemblywoman think it should do so)? (The whereases are supposed to function as reasons, but, even though not all of them are quite factual, was there some specifically New Jerseyan problem that they thought needed to be solved? That it should be taught in schools, alongside Greek and Latin?)
Another odd claim is that Panini's Ashtadhyayi "removed the need for Sanskrit's evolution".
Y said,
July 25, 2025 @ 9:40 pm
Is Prof. Liberman's puzzlement sincere or ironic?
It seems pretty obvious that the resolution was composed by an ignorant Hindu nationalist, who fed it to ignorant gullible American lawmakers, who think they are merely honoring an immigrant culture. The "not created by man" probably refers to some ostensible proof of Sanskrit's divine origin.
(Which are the benighted 3% of the world's languages not "influenced" by Sanskrit? How is Sanskrit their mother, as of all the world's languages, yet did not influence them?)
Barbara Phillips Long said,
July 25, 2025 @ 9:48 pm
For those who aren't familiar with New Jersey or its Indian population — which may have influenced this legislative proposal — Wikipedia has more information:
The U.S. state of New Jersey, where most of the population is situated within the New York City metropolitan region, has by a significant margin the highest proportional Indian population concentration of any U.S. state. According to Census estimates in 2023, 4.6% of New Jersey's population consists of individuals of Indian origin. New Jersey is the state with the highest percentage of individuals with Indian ancestry in the United States, at over 5%.
https://en.wikipedia.org/wiki/Indians_in_the_New_York_metropolitan_area
Jonathan Smith said,
July 25, 2025 @ 10:09 pm
Hmm, this seems to part of an old and ill-conceived Indian nativist/nationalist soft-power play, call it "RickBriggs". Whose language use is shall we say distinctive and whose publications are shall we say not extensive. I could be projecting based on the contemporary HanChauvinist gimmix with which I'm more familiar of course… but probably not.
AntC said,
July 25, 2025 @ 10:17 pm
AFAICT this slop was generated without the assistance of AI, which just goes to show … something.
Jonathan Smith said,
July 25, 2025 @ 10:51 pm
re: (part of) the linguistic part, where the concern is ostensibly efficiency or non-ambiguity or sth., Classical Sanskrit is pounded into the pavement by (among others) Japanese, where e.g. (extracting Sanskrit+English from this useful description by Vikram Chandra) —
Sanskrit Rāmaḥ hastena brāhmaṇāya dhanam dadāti: "Rama gives money to a Brahmin with (his) hands"
becomes by online translation consensus
Japanese Rāma wa baramon ni te de okane o ataeru
i.e. a literal list of four arguments straight from the lexicon + role labels straight from the lexicon + verb where ok admittedly some tiny number of Pāṇini-like derivational rules are indeed called for.. but 99% of it, most notably (in modern argot) phonology>phonetics, can go right in the trash, saving yer "AI" project bowcoo CPU cycles.
AntC said,
July 26, 2025 @ 2:13 am
Wikip seems to think "the ancient Dravidian languages influenced Sanskrit's phonology and syntax.", which at least refutes the 'oldest language' claim; and rather weakens the 'influence'.
Peter Grubtal said,
July 26, 2025 @ 3:34 am
You don't need to be an authority on Sanskrit (which I'm certainly not) to recognize this as trash.
I don't want to be unfair to our US cousins, especially since I have a poor understanding of how the state legislatures operate over there, but this seems to be on a par with the attempt to legislate for the value of pi.
Gokul Madhavan said,
July 26, 2025 @ 4:54 am
For those interested in learning a bit more about the Indian tradition of “knowledge representation”, I'll provide a short overview as this is something close to my heart (and to my dissertation). The Sanskrit term for what Rick Briggs is talking about is śābda-bodha, literally “language-related cognition”. While debates in Indian philosophy about the kinds of mental structures invoked by the use of language are extremely ancient, discussions of śābda-bodha specifically are a product of the last few centuries, following the Epistemological Turn in Indian philosophy prompted by the magisterial Tattva-cintāmaṇi of Gaṅgeśa (14th c.). From that point on, Indian philosophers belonging to several different Hindu traditions (the Buddhists having mostly departed the scene) all developed increasingly sophisticated accounts of the structure of linguistically generated cognitions. There were three major rival philosophical positions here, each one articulated by a different philosophical tradition, and each one according primacy to a different linguistic element:
the Naiyāyika position, in which the entity signified by the word with the nominative case-ending is the primary qualificand;
the Mīmāṃsaka position, in which the verbal force signified by the verbal ending is the primary qualificand; and
the Grammarians’ position, in which the action signified by the verbal root is the primary qualificand.
Each of these theories has its own strengths and weaknesses, which we don't need to get into here. The main point I'd make here is that it is somewhat misleading to think of this as “Pāṇinian analysis”. Pāṇini lived around two millennia prior to any of these philosophers, and while they were all deeply and intimately familiar with the Aṣṭādhyāyī and used its terminology and techniques liberally, they were operating at a different level of abstraction from Pāṇini.
Gokul Madhavan said,
July 26, 2025 @ 5:39 am
To clarify, let's take Jonathan Smith’s example sentence.
The sentence Rāmaḥ brāhmaṇāya hastena dhanam dadāti would be analyzed into its component morphemes by the Pāṇinian system as Rāma-sU brāhmaṇa-Ṅe hasta-Ṭā dhana-am (dā-ŚLU)-tiP, where the bolded morphemes are Pāṇini’s abstract case-endinags. We can indeed see a one-to-one match between the Sanskrit and the Japanese at least for the nouns (though I'd say that Rāma-ga would be a closer literal translation of the Sanskrit grammatical suffix here, but that’s neither here nor there).
Now this in and of itself is a pretty formidable achievement, as it manages to peel back Sanskrit’s rather opaque surface structure to identify the underlying morphemes. The synonoymous sentence Dāśarathiḥ agrajanmane pāṇinā vasu yacchati, although sharing virtually nothing in common with the previous sentence, ends up having a virtually identical Pāṇinian parsing (Dāśarathi-sU agrajanman-Ṅe pāṇi-Ṭā vasu-am (yam-ŚaP)-tiP).
But the Pāṇinian system extends beyond morphosyntax into semantics through its postulation of kārakas (sometimes translated as “theta roles” though the correspondence is inexact). We would thus be able to encode Rāma as the KARTṚ (“agent”), the brahmin as the SAMPRADĀNA (“beneficiary”), and so on, based on the Aṣṭādhyāyī itself.
This is where Pāṇini ended and where śābda-bodha analysis began. Scholars developed different signification functions to capture the myriad ways in which words can refer to entities. (Thus, our two synonymous sentences would be captured as having the same referents while still using different words to refer to them.) They found ways to connect these entities to one another via kāraka relations and other relations, and use these to eventually construct elaborate networks that interweave signification, syntax, morphology, and ontology as well.
To express all of these patterns verbally, these scholars utilized the technical jargon of early modern Indian logic, the so-called Navya Nyāya. A full-blown śābda-bodha analysis of a regular Sanskrit sentence with a couple of compounds and a subordinate clause or two would easily be a dense, paragraph-long labyrinth in the technical jargon of this discipline. (Even contemporary Sanskrit scholars who were not trained in this jargon found this stuff hard to swallow, and we have examples of verses making fun of these scholars in some literary texts.) Translating such texts into English would be virtually incomprehensible, but drawing them out as directed graphs makes them much easier to grasp.
This sort of semantic analysis was actively applied to solve problems in Sanskrit literary theory (when trying to redefine and reclassify figures of meaning) and legal theory as well (when interpreting older texts on property and inheritance and so on).
AntC said,
July 26, 2025 @ 8:03 am
For those interested in learning a bit more about the Indian tradition of “knowledge representation”, …
The first requirement for knowledge is to distinguish facts/truth from … trash, to echo m'learned colleague above. If this resolution is a product of Indian whatever, it's showing that tradition as worthless. ( I believe Sanskrit learning to be considerably better than that.)
OTOH perhaps we should cut the dude some slack: the 'scientific evidence might be bogus, but at least it's having no impact on others. Not, for example, undoing two generations of vaccine protection/public health.
languagehat said,
July 26, 2025 @ 8:04 am
Y said all that needs to be said succinctly and well; I'll just add that the "Indian tradition of knowledge representation” is irrelevant here: this is a story about idiotic local politics and tendentious ideological PR, not Sanskrit itself. I join Y in puzzlement at Prof. Liberman's appearing to take this in any way seriously rather than mocking it as it deserves.
Lightning Ghost said,
July 26, 2025 @ 8:55 am
What Y said—I would add (from a South Asian perspective) that this set of nonsensical claims about Sanskrit is a well-established and commonplace fixture of Hindutva discourse.
Richard Hershberger said,
July 26, 2025 @ 11:58 am
To answer why the New Jersey legislature might take something like this up, the key is that it will have absolutely no practical outcome. Some legislators judge that some of their constituents will have warm fuzzy thoughts come election day if they vote for it, while not in turn making other constituents angry. So it is a win.
This sort of thing is pretty common. For an example from sports history, the standard (and almost certainly true) origin story of basketball is that it was invented by James Naismith in Springfield, Massachusetts in 1891. There is a lot of background context to this, but that is the bare bones. In the mid-20th century a competing claim arose, that the game was invented by one Lambert Will in Herkimer, New York, in 1890, and the Naismith stole the idea from him. Since then Big Basketball has engaged in a vast conspiracy to hide the truth. The evidence for any of this is most generously described as "tenuous." But in recent years an activist group organized, and they persuaded the New York State legislature to enact a resolution saying it is totally true. Why did the legislature do this? So members of the group would be more likely to vote for them. Why did the group do this? Because now they can proclaim their version to have been officially confirmed. Only the gullible would find this persuasive, but that is a large swath of humanity, and clearly the target audience for this stuff.
Stephen Goranson said,
July 26, 2025 @ 12:02 pm
Some have claimed that Hebrew was the original language.
Or Irish…
DaveK said,
July 26, 2025 @ 2:03 pm
Richard Hershberger is correct about this resolution being a species of nonsense that’s endemic in state legislatures. The smoking gun is the instruction that a copy of the resolution be sent to Samskrita Bharati Nee Jersey. This organization’s website states:
We are a volunteer-driven organization that conducts Sanskrit classes for children and adults.
Our Centers (Kendrams) are located in Piscataway, Princeton, and Parsippany.
Some members of the state Senate and Assembly wanted to do them a favor and copied the group’s handout into the text of the resolution without a critical reading.
Mark Liberman said,
July 26, 2025 @ 2:26 pm
@Sptehen Goranson: "Some have claimed that Hebrew was the original language":
See "Edenics", 11/1/2013.
MattF said,
July 26, 2025 @ 3:06 pm
@Peter Grubtal
Well, if the BITM can legislate an exact value for the speed of light (in SI units) why can’t the great state of Indiana legislate a value of pi? Seems… unfair, somehow.
MattF said,
July 26, 2025 @ 3:12 pm
BIPM, I mean.