Legion, thanks for your in-depth response. I don't think there's anything mysterious at all, I think I've a pretty good idea of how these processes work. What I'm raising the point for is as a talking point with those who disagree with the Aryan migration theory, who generally claim that there have been no migrations of Indo-Europeans into India.
Ah! Makes perfect sense. Thanks. I have come across that view here once or twice before, and to be honest I don't understand it (it's similar to the idea that the Syriac gospels were written in the Aramaic of Jesus). Ironically, there is a connection between
The White Goddess and one theory of the origins of the "Indo-Europeans" in that Graves' text was highly influential for certain strands of thoughts as well as the development of Wiccan traditions and Goddess spiritualties, and the Kurgan hypothesis which is still accepted by many was proposed and defended by Marija Gimbutas in the 70s, before she abandoned reputable scholarship for a (pretty much baseless) theory of matriarchal prehistory.
Theories tend to fall into two categories (although there are no shortage of minority views that are still reasonable, compared to the "Sanskrit primacy" idea). Put more simply than I could:
"Currently, there are two types of models that enjoy significant international currency.
There is the Neolithic model that involves a wave of advance from Anatolia c. 7000 bc and, at least for south-eastern and central Europe, argues primarily for the importation of a new language by an ever growing population of farmers. This part of the model has reasonable archaeological support in that there is a fair amount of archaeologically informed consensus that derives the earliest farming communities in the Balkans from somewhat earlier farming communities in Anatolia. For the periphery of Europe the means of explanation become less clear, and rather than a language expansion driven primarily by Early Neolithic population expansion, this model now seems to admit of later (Late Neolithic, Bronze, or Iron Age) movements into Mediterranean, Atlantic, and northern Europe. For the steppelands, it envisages the spread of an agricultural economy from the Balkans to the steppes where it was then carried, in the Bronze Age, beyond the Urals and then south into the territories of the historic Indo-Iranians and Tocharians. Some opponents of this solution admit that the initial archaeological scenario may be true but suggest that the Early Neolithic farmers spoke an unknown non-Indo-European language, possibly related to the historically attested non-Indo-European languages of Anatolia (e.g. Hattic, or possibly one of the Caucasian languages).
Alternatively, there is the steppe or kurgan model which sees the Proto-Indo-Europeans emerging out of local communities in the forest-steppe of the Ukraine and south Russia. Expansion westwards is initiated c. 4000 bc by the spread from the forest-steppe of mobile communities who employed the horse and, within the same millennium, wheeled vehicles. These intruded into southeastern Europe at a time when there was major restructuring of local societies (variously attributed to climatic change, local social evolution, or intrusive steppe populations or a combination of the three). The hard archaeological evidence, i.e. the recurrence of the classic steppe burial type in the Balkans, is reasonably solid as far as the river Tisza. Beyond Hungary, this model relies on far less stringent archaeological evidence. A central component is that it requires some form of genetic derivation of the Corded Ware culture of the north European plain from the steppe cultures (one can talk either of direct derivation or the spread of a symbolic and social system that was initiated in the steppe). As for the Asiatic Indo-Europeans, it offers the model that was adopted later by those who support the Neolithic model. Opponents of this theory would tend to see the steppe cultures as the ancestors of the Indo-Iranians and possibly the Tocharians but not of the entire Indo-European family."
Mallory, J. P., & Adams, D. Q. (2006).
The Oxford Introduction to Proto-Indo-European and the Proto-Indo-European World. Oxford University Press.
Such claims are often associated with beliefs that Sanskrit originated in India, and that all Indian languages, including Dravidian ones, descend from Sanskrit.
I wasn't aware that proponents of such claims were aware of Dravidian languages. That's an improvement!