The
Joke That is “AI”: The Evidence of the Indo-European Numbers
Shrikant G. Talageri
I have already pointed out the utter stupidity, inanity and un-intelligent nature of AI when it comes to analyzing issues involving the “Aryan” problem. Here is a classic example of this stupidity, inanity and un-intelligent nature:
A reader made the following comment on my last article. He should have made this comment on my article on the numerals, but nevertheless I am grateful to him for bringing this idiocy to my notice and thereby enabling me to give a fitting reply to it and to expose the utter uselessness of “AI” in these matters:
“I ask AI to analyze your work titled "The Irrefutable Evidence of the Indo-European Numbers". What do you make of this?
The Flaw: The "Regularity" Paradox Where Talageri’s argument departs
from scientific consensus is his underlying assumption: Sanskrit is perfectly
regular, therefore it is the oldest.
In
historical linguistics, this assumption is practically reversed. This brings us
to a foundational concept called The Principle of Archaic Irregularity.
Irregularity is Ancient; Regularity is Invented. Language naturally creates
messy, irregular fossils over thousands of years. When you see a language with
a perfectly logical, seamless grammatical or numerical system, it is almost
always because the speakers (or grammarians) regularized it at a later date to
make it mathematically consistent.
Think of English verbs: Play/played is a regularized, newer invention. Go/went
or sing/sang/sung are chaotic, irregular fossils that survived from thousands
of years ago.
English eleven and twelve come from the ancient Germanic roots ain-lif
("one left over ten") and twa-lif ("two left over ten").
They are weird, isolated, and irregular—which tells linguists they are
incredibly old, localized Germanic innovations, not a sign of a decayed
Sanskrit system.
Sanskrit’s numbering system is brilliant, but its flawless consistency is the
hallmark of a language that was actively streamlined and polished (which, as
discussed previously, is exactly what the ancient Indian grammarians did). It
is a perfected system, not a raw, primitive one.
The Steppe Model Fits the Same Data. Talageri argues that the European base-20
contamination proves the languages traveled from India to Europe. However, the
mainstream Steppe Homeland model accounts for the exact same data without
breaking the rules of linguistic evolution.
If
the homeland was on the Eurasian Steppe:
1. The Steppe nomads had a pure decimal (base-10) system.
2.The branch that migrated south into India (Indo-Aryans) kept the base-10
system, expanded it, and eventually polished it into the perfect, infinite
mathematical series seen in Sanskrit.
3. The branches that migrated west into Europe collided with indigenous
European hunter-gatherers and early farmers who counted in twenties
(vigesimal). The migrating IE speakers absorbed this local habit, creating the
hybrid base-10/base-20 systems we see in Celtic and French today.
The Verdict
Talageri’s
data collection is excellent. He correctly identifies that European IE
languages absorbed foreign counting systems, while Sanskrit retained a pure
decimal framework.
However, his conclusion relies on the idea that "perfectly regular"
equals "oldest." By the established metrics of historical
linguistics, a flawlessly logical system like Sanskrit's numbers is the result
of later refinement, while the messy irregularities found in European branches
are the ancient, unpolished fossils of tribal migrations.
The data he presents is completely valid, but it does not uniquely prove an
Indian homeland—it simply proves that ancient Europe was full of
non-Indo-European people who counted in twenties.”
The reader will note
that this utterly un-intelligent “AI” seems to have not
even glanced at, let alone read in detail and analyzed, my articles on the
Evidence of the IE Numbers. This un-intelligent “AI” is
not replying to my article, but to all those Indians who claim that Sanskrit is
the most perfect language for computers. And then it is juxtaposing that claim
to its “verdict” on my article on the IE numbers without paying any heed to
what is written in that article, almost as if it has not “read” that
article at all.
Note the two fundamental
fallacies that AI un-intelligently assumes that I have
made in my article
Fallacy 1: “Most Regular is Oldest”.
To begin with, please
someone should start out by pointing out where in the article have I expressed,
even indirectly, the “underlying
assumption: Sanskrit is perfectly regular, therefore it is the oldest”,
or “However, his conclusion relies on the idea
that "perfectly regular" equals "oldest." By the
established metrics of historical linguistics, a flawlessly logical system like
Sanskrit's numbers is the result of later refinement, while the messy
irregularities found in European branches are the ancient, unpolished fossils
of tribal migrations.”.
This “AI” preaches to
me (which, in this particular instance, is like “carrying coal to Newcastle”): “In historical linguistics, this assumption is practically
reversed. This brings us to a foundational concept called The Principle of
Archaic Irregularity.
Irregularity
is Ancient; Regularity is Invented. Language naturally creates messy, irregular
fossils over thousands of years. When you see a language with a perfectly
logical, seamless grammatical or numerical system, it is almost always because
the speakers (or grammarians) regularized it at a later date to make it
mathematically consistent.”
Not only have I nowhere
claimed that Sanskrit has the most perfect and regular number system (let alone
used that as an argument for it being the oldest), I have constantly,
throughout the article, pointed out places where Sanskrit has irregularities.
or irregular features, which have been regularized and improved in different
other later IE branches and languages:
1. The practice of
placing the unit numbers before the tens number (catur-viṁśati,
instead of viṁśati-catur, for 24, which, continuing in all
Indo-Aryan languages, leads to confusion; whereas many other IE languages have
reversed the order).
2 . Having a
minus-principle (ūna-triṁśat, for 29, instead of viṁśati-nava,
which, again, has been corrected in most other IE branches and languages).
3. Having irregular
forms because of its rules of sandhi. (a characteristic which has
resulted in extreme irregularities in the later IA numbers, even when those
late IA languages don’t have rules of sandhi, and making the
modern IA languages of North India the most irregular and difficult
numbers in the world).
The simple fact that I
place Sanskrit, along with Tocharian and spoken Sinhalese,
in an earlier stage of development (and only because it does not form
the numbers 11-19 in a different way from later sets like 21-29, 31-39, etc),
does not mean that I am implying that Sanskrit is the most regular
(let alone that this makes it “older”) – regularity and oldness are not
issues at all – and in fact even a child should be able to see that, within that
stage, Tocharian and spoken Sinhalese are more “regular”
than Sanskrit.
If this “AI” is not
able to grasp even this basic point, I am frankly speechless.
Fallacy 2: “Sanskrit has a Pure Decimal System, European IE
Languages are 20-infiluenced”.
“AI” fatuously writes:
“The branches that migrated west into Europe collided with
indigenous European hunter-gatherers and early farmers who counted in twenties
(vigesimal). The migrating IE speakers absorbed this local habit, creating the
hybrid base-10/base-20 systems we see in Celtic and French today.”
“He correctly identifies that European IE languages
absorbed foreign counting systems, while Sanskrit retained a pure decimal
framework.”
“The data he presents is completely valid, but it does not
uniquely prove an Indian homeland—it simply proves that ancient Europe was full
of non-Indo-European people who counted in twenties.”
This un-intelligent
“AI” totally fails to understand that I have nowhere made these claims. Nowhere
have I claimed that Sanskrit has the most perfect decimal system, while European
IE languages alone have been influenced by the vigesimal (20-based)
number system.
1. All the
later IE languages of stage 3 and 4 (other than Sanskrit, Tocharian
and spoken Sinhalese have been influenced by the vigesimal
effect) including the Iranian
languages and the later Indo-Aryan languages in India.
[Surely, even this un-intelligent “AI” understands that Sanskrit
is older than the modern IA languages, and that therefore
the vigesimal effect is a newer phenomenon as compared to an older
decimal system!]
2. I have pointed out
in detail that it is not only in Europe but in India
itself that there are many vigesimal (20-based) languages in every
corner of India which caused a vigesimal effect on the IE languages in
the third stage.
3. The vigesimal
effect, which this “AI” calls “the hybrid
base-10/base-20 systems we see in Celtic and French today” has definitely
come about due to a very late effect because “ancient
Europe was full of non-Indo-European people who counted in twenties”.
This is much later, and a completely different
effect from the vigesimal effect which is found in all the IE
branches (other than the earlier named three languages, and therefore
presumably also other than Proto-IE and Anatolian), and is
just a later aberration which has to be pointed out but which plays no part in
the general process of the development of the numbers in other IE languages.
In short, “AI” flops
again.
[Incidentally, spoken Sinhalese is in the second stage, while Literary Sinhalese adopted the third stage from the Prakrits. But it simplified the irregularities, and therefore, even as it represents the earliest recorded stage, it represents a regularized developed form over the ages]
