Years ago I met a young man from Greece. He introduced himself as ‘Petros”, but what surprised me was the way he pronounced his own name. He said, “My name is Path-rr-os”. The way he said it, the name sounded like the Hindi word for rock, “patthar”. In fact, Petros means “stone” or “rock”. The word “petrification” in English is related to that, and the name Peter is a variant of Petros. Travelling south from England to Italy, one asks for hotel room in England. In France, one asks for a “chambre”, pronounced shom-bruh, which means chamber or room in English. When one crosses the border from the south of France into Italy one must ask for a “camera” when looking for a room. “Camera” in the Italian language means room or chamber, but the Italians pronounce it “kamra” which is exactly the same as the word for room in Hindi. Sanskrit “danta” for tooth becomes “daant” in Hindi, to be treated by a dent-ist in English. The word father is derived from Latin “pater”. That is remarkably similar to Sanskrit “pitru”, from which we get Hindi “pita”. The list of such words that seem to have a common origin in the ancient past is almost endless. Any Indian who knows English in addition to one or more Indian languages may already know that English, Italian, French, Russian and Indian languages share many similar words.
However Europeans did not know this until they came to India and “discovered” Sanskrit. Sanskrit provided the missing link that made people understand relationships between a huge group of languages spoken from India to Europe. Sanskrit, especially old, Vedic Sanskrit has more words in common with more European languages than any other language including Greek and Latin. The science (or pseudoscience as it turns out to be in practice) of comparative linguistics was born with the western “discovery” of Sanskrit as described in a eulogy by Max Muller. More than a century earlier than Max Muller, a Briton called William Jones in a much quoted speech had theorized that Sanskrit, Greek and Latin must have sprung from a common root language. Comparative linguistics sought to rediscover that language.
Comparative linguistics started initially with similar sounding words in different languages with similar meanings (called cognates) but as time went on, in their search for a “mother language” linguists developed hundreds of rules to decide whether words were actually related or not. The grammar and the “form” of the words had to be declared right. Linguists picked up dozens of European languages and looked for parts of words and grammatical structures that were similar, sometimes rejecting obviously similar words as unrelated or “borrowed” and sometimes including unlikely sounding words as cognates based on their rules. Rules were created and when those rules did not work they created new rules and still more rules in their quest for the holy grail of mother language uniting all languages, but there is no evidence that they even once applied scientific method to linguistics.
The most obvious slip up is the assumption that all modern languages must have descended in a tree like fashion from one mother language. That is to say that there is one trunk, dividing into branches, smaller branches and leaves. In fact things might not have panned out in this manner. There may have been a “trunk” mother language that branched out into daughter languages but those branches, unlike real tree branches may have coalesced to form one branch and then re-branched. How would this happen? Real life human communities in the past may have moved to different places as populations grew, where their mother language evolved, but epidemics, drought or floods would likely have driven back many to re-mix with their original communities, and the two slightly evolved sister languages would then have mixed in an unpredictable fashion. But linguists don’t seem to have accounted for this and seem to favour a relentless outward spread of language from one point – particularly favouring a theory that involves conquest and domination based on technology. Having only one unprovable theory to explain the unknown without considering other possibilities is as unscientific as any study can get. The ease with which linguists have assumed language relationships and the eagerness with which they have developed new rules to explain anomalies has a good analogy in the history of science.
In the 18th century “scientists” in Europe believed that objects burned because they contained a substance called “Phlogiston” (pronounced flojiston). The phlogiston was released upon burning leaving the object lighter (ash). When someone noticed that objects would not burn fully when placed in a sealed jar with air cut off, they said that air could hold only so much phlogiston and that burning stopped when air got overloaded with phlogiston. When someone else pointed out that some metals burned and the end result was heavier and not lighter than before, the scientists said that in those cases phlogiston had negative weight – so removing it caused the burnt object to become heavier. In an eerily parallel manner linguists have not attempted any scientific method to resolve anomalies and have assumed from the word go that their reconstructions of an old, unknown language from cognate words would work as long as they muddled along making up rules at every roadblock.
What scientific method could linguists have applied in language reconstruction? Taking a cue from any experimental research, one would know that a man trying to build an aircraft would first build a model and see if the theory that an aircraft can fly would work. And if the model works, it can be used as the basis for building the real thing, or alternatively, used to discover what might be wrong if it did not work. There is no evidence that linguists did that. They could simply have taken three or four modern languages with a known old ancestral language and then tried to reconstruct that old language from the modern ones. If their methods reproduced the old language at least in terms of 60% of the words – they could assign 60% accuracy to their reconstructions. For example they could have taken Hindi, Gujarati, Marathi and Bengali and got a group of linguists to follow their rules to reconstruct Sanskrit, a known older language. If there was success the method would be validated. But even if it failed, it would give information on the routes to be followed to find out why the reconstruction was wrong. This has never been done. Linguists have just rushed in and merrily reconstructed unknown languages as if all the methods used are valid and correct. And verification is impossible because the reconstructed language is an imaginary one. On top of this they have declared the reconstructed language as having really existed and sometimes used it as a model to reconstruct an even older language. An example of this is Avestan, an imaginary language that Parsis are said to have spoken in 1000 BC. Avestan has been reconstructed from fragments of original gathas recalled orally, from Pahlavi language texts translated into Sanskrit. Linguistic references admit that translation and reconstruction from Pahlavi texts is “fraught with difficulty”. With all these handicaps, the Avestan language has been reconstructed and declared to have been a real language 3000 years ago (covered in more detail here).
Now we move on to the way linguists have gone on to “reconstruct” the “mother language” of many Indian and European languages and all languages in between. This has been given the name “Proto-Indo-European” or P.I.E. (also PIE) for short. Linguists have taken a list of 200 or so cognate words from over 2 dozen “Indo-European” languages to work backwards and reconstruct PIE. This might be an interesting exercise so long as everyone is clear that the new language is simply a model and not a real language. But linguists seem to think that they have reconstructed a “real language”. Linguistics references constantly claim that a word in say English, or German or Hindi sounds (or means) what it does because it had an ancestral PIE word that they cite as a real word. This is a circular argument. If you build a language using say 26 other languages, that “built” language will obviously bear a relationship to words in one or more of those 26 languages. No surprise there. A cheese sandwich is a cheese sandwich because it contains cheese, bread and butter. Those three constituents were not created out of the sandwich. There is a scientific instrument called a “mass spectrometer” that detects the component atoms of any substance placed in it. If you place salt in it, it will detect sodium and chlorine. This is because common salt is made up of sodium and chlorine, not because salt somehow “created” the two components the way PIE is supposed to have given rise to the languages that it was made up of in the first place.
I performed a little experiment using a PIE wordlist. I imported a list of over 100 cognate words from 34 Indo European languages into a spreadsheet. Such a wordlist has been used to reconstruct PIE and it consisted of words for body parts, kinship, terrain and a few other common word types that one would expect in any language. A maximum percentage (78%) of the cognate words were present in Sanskrit, 76% in Latin and 72% in ancient Greek. This is not surprising or odd. After all PIE was built up from many languages and Sanskrit, Greek and Latin are the biggest contributors to that PIE word list. But what is really odd about that word list is that they have included outlier languages like Luwian, Thracian, Ossetian, Umbrian, Lydian, Phrygian, Tocharian and Hittite. Many of these languages are practically unknown and are related to just 1-10 words in that 100 plus word PIE list. Languages that have been “reconstructed” or whose evidence has been found via Sanskrit texts such as Avestan and Tocharian score higher. A list of 34 languages in which one-third are “outliers” which have been deliberately included despite their weak contribution to the wordlist suggests that there is a contrived effort to “force fit” outliers into a larger group just to show how PIE is mother to all these languages despite very little being known about them. If those languages had large contributions from non-IE languages we will not and cannot know, but they have been bundled into the IE group anyway. Empirically there appear to be better choices that could be made to study languages rather than force fitting them into a predetermined theory.
Finally, after creating the imaginary language, PIE – linguists have washed their hands off, saying that their job is done and that they will leave this gift to archaeologists to fix a date and place. And PIE has dropped like manna from heaven for archaeologists who have gleefully grabbed and dumped PIE in a place where there is no historical evidence of language, the Russian steppe – a convenient area in the middle of a land mass equidistant from every corner making it seem “likely” that the language could have gone anywhere from there because no place can be declared as “extra far away” from that point. “Too far away” has actually been used as an excuse to rule out certain other possible places of origin. Needless to say, archaeological findings have been cooked up to connect India with that imagined place of origin of the imaginary language PIE – but that would be a topic for another day!
An incurable patriot, Dr. Shivsankar Sastry is a surgeon by profession; and a historian, thinker, sociologist and military aviation enthusiast by choice