Dear List,
In the Amarakosha, a thesaurus of Sanskrit darting from the 5th-6th century CE, the word cakra 'wheel' is explicitly stated to also mean `an army'. Monier-Williams' Sanskrit-English dictionary, too, says cakra is attested with the meaning `troop of soldiers, army, host' in the Mahabharata. Note I am here talking of the *uncompounded* word cakra, not of the compound cakravyUha `circular array of troops', which famously occurs in the Mahabharata.
This lexical equation cakra = army leads me to speculate a bit. Could this specialized meaning of the word cakra be traced back to Indo-European times? Let us see if the semantics of PIE *kWekWlo- can suggest some reply to this question.
The nominal stem *kWe-kWl-o- `wheel' (lit. `turn-turning', an iterative or reduplicated word-formation), derived from the PIE root *kWelh1- `to turn, revolve, move round, dwell round, go around or back and forth', and possibly also suggesting the cyclic, wandering movements attributed to pastoralists, may have existed in the protolanguage with meanings other than `wheel' -- for instance, with the meanings `circle, circumference' and `(the folk) wandering around / dwelling round'. Linguist Raimo Anttila (see his "Note on Umbrian poplo" at
http://tinyurl.com/d32fsv3 ) puts forward the hypothesis that one of the variant meanings of *kWekWlo- may have been `group of people in arms' in the sense of `troop of people going/travelling around'. In some daughter branches of IE the root *kWelh1- is central in conveying all kinds of going around, e.g. as servants, shepherds, etc. This linguistic hypothesis is based on the etymologizing of Osco-Umbrian poplo- ~ puplu- as a derivative of Proto-Italic *poplo- `army, group of people/citizens in arms', in turn, supposedly a reflex of PIE *kWekWlo-. Latin populus (earliest attested form: poplos) `human community, people' is considered by a number of linguists and philologists to be a loan from Osco-Umbrian (probably Sabine) poplo- `citizens in arms'. Similarly, the Latin deponent verb populari `to devastate, ravage, plunder' can be understood as a specialized military sense of `travel around', that is, `to have an army pass through'. Although Latin is a "Q-Italic" language (i.e., its reflex of PIE *kW is *k), it appears to have borrowed a number of basic words from neighboring "P-Italic" languages (i.e., languages in which the reflex of PIE *kW is *p like in Proto-Italic itself) such as Sabine and Oscan; among such likely borrowings, let me recall bos `ox', lupus `wolf', and popina `cook-house'.
With this in the background, let us now come back to Sanskrit cakra. Could not the specialized meaning `army' of this word be a reflex of PIE *kWekWlo-, just like Osco-Umbrian poplo `group of people in arms'? (Cf. Hartmut Scharfe's footnote 213 on p. 52 at
http://tinyurl.com/cf4hyht )
Kindest regards,
Francesco