William Jones wrote a book, The Sanscrit Language, to tell that Greek and Latin had a common root with Sanskrit, and there must have been a Proto-Indo-European language from which to derive human speech.
Mr. Jones learned Greek, Latin, Persian, Arabic, Hebrew, and basic Chinese, says Wikipedia to sum up that he knew 13 languages thoroughly, and another 28 reasonably well; in total, some 41 languages.
His Proto-Indo-European, PIE as a cookie in short, was to have given origin to European tongues as English or Polish.
All these languages would have to share words from the same source. Do they have a root PIE vocabulary? A reasonably good knowledge of a language should encompass words as woman, man, child, and house. Let us compare these words in English, Latin, Greek, Russian, Polish, German, French, and Sanskrit.
I do not know Sanskrit. I can only compare resources. The morpheme –man, quoted by supporters of the PIE, yet refers to thinking and not to gender, whereas it is common lore that masculinity is not strictly synonymous with pensiveness.
Sanskrit resulted from conditions of rigid social stratification for status and ancestry, says Wikipedia. Even words for children became different with regard to heirloom, and “children of men” were not sutah; they were given another name, napraja. We could not say, let us look for a common root with the PIE, because it must have been a beautiful culture.
Vir or andros, child or riebienok, woman or kobieta ― the words do not resemble one another, and they are the basic vocabulary that hardly ever changes. This is the vocabulary to compare for language grouping. Groups can work better than “families”, as “language families” derive tongues one from another.
Polish and Russian could make a group. We may compare the words muzshtschina and mężczyzna. There is not much point deriving Polish from Russian or Russian from Polish, however. We can compare riebienok and dziecko, or zshenshtschina and kobieta.
The Proto-Indo-European theory yet would insist to derive, and even create theoretical word shapes, as if there might have been some ancient wizard tongue.
Oldest does not mean wisest
Proto-languages are constructs purposed at a theoretical “family”, whereas natural language does not provide any rationale for probabilistic forms of speech. Machine or otherwise built, a code is not a natural language, and word shapes are not forgotten cousins.
Taking the Russian and Polish words for children, to have a “precedent” for both dziecko and riebienok, we would have to make up a “proto-word”. Even if we created a shape as *dieriebko, the form would not evidence that such a word ever existed, and honestly, it is unlikely.
I do not share in the enthusiasm or fascination with speculative content. Quotient always has been a matter by a human individual (also when intelligent people meet up), and neither people nor languages were wiser in ancient times.
Evolutionary frameworks are more plausible: languages might have emerged on Earth in result of human cognitive progress. People also shared knowledge, and similarities as domus or dom, as well as house or Haus evidence urban progress and contacts; they are not used in language grouping.
Sanskrit might have been influenced
Languages do not develop in isolation, and Marco Polo was not probably the first visitor to the Far East. Let us think there is language A. Some people come around and adopt grammar for verbs from A, even in a considerable extent. They do not refer to language A for everything, however. They already have a language, B.
The people who talk B make progress; they begin to come up with new words, and language A begins to adopt from language B: if we find a phrase or word in Sanskrit today, it does not mean the wording has been there since the beginning of time.
If to have Russian, Polish, English, or German for kindred, it would be by marriage. English has four verb Aspects. German vernacular regiolects purportedly do have forms that correspond with the Progressive, but they are condemned by standard. Russian or Polish also would remain by one Aspect, and the cognitive variables by other Aspects are provided for by speech parts other than the verb.
English: I’m reading.
Polish: Teraz czytam.
Everyday and natural language lets the context work:
English: What are you doing? ~ Reading.
Whoever to ask, who knows Polish, if czytam maybe means here that it is usually or always, he or she will naturally deny. The matter will be yet the same, if to ask whether the sense would be rarely or hardly ever. The cognitive reality is in Polish the same as in a person to say in English, I’m reading. The difference is in language forms.
Evidence and dating
Decent linguistic work requires evidence. The Rosetta Stone was absolutely unique; still, it covered only the Ancient Egyptian ― the glyphs and the demotic ― along with Ancient Greek. It did not have etymologies.
In short, the best source for comparative linguistic evidence in human history so far, the Rosetta Stone allowed translation, but not word derivation. There never was even a stone like the Rosetta, for “Indo-European” languages.
Rosetta granodiorite material would not allow carbon dating. Researchers have relied on dates inscribed in the stone. Where possible, linguistic evidence should be carbon-dated. There is no philological method to affirm on originality beyond evidence. Writings were copied in Antiquity, the Middle Ages and later, hand style and pen craft.
There is no C14 or stone evidence for the Proto-Indo-European. A few researchers have claimed evidence, but their data are questioned for poor quality (Wikipedia, PIE history of research). Supporters of the “Proto-Indo-European family” have yet gone into making the Proto-Indo-European religion. There is no PIE root for planet Earth and life.
It seems there was a pie rather than the the PIE, in the times of Mr. Jones, and that pie was the Company rule in India. The British colonial rule in India began around 1500, and there was much rivalry.
Space 1999 would accurately show Proto-Sanskrit reading…