...Just as the cultures of the steppes have various historical origins and not solely Pontic, similarly, the BMAC has predecessors which are archaeol… - Jean-Paul Demoule

While no one would dream of questioning the resemblances between the various so-called Indo-European languages, the centrifugal arborescent model in its current forms cannot be considered as validated due to the numerous contradictions that it contains. Furthermore, abuses, both past and present, of this model should incite us to the utmost rigor. We must therefore turn toward much more complex and multidisciplinary models concerning historical phenomena that span millennia if we are to meaningfully explore the multiplicity of problems that make up the “Indo-European question.”

The model of diffusion of “archaeological cultures,” each corresponding to a given “people,” is both naturalist and directly inspired by the nation states of the nineteenth century; it does not correspond to numerous situations provided, for example, by the ethnology of other continents or by the history and archaeology of protohistoric peoples of Early Medieval Europe.

At least one thing is sure: the collapse of both these urban civilizations (i.e., those of the BMAC and the Indus) was not caused by attacks by Andronovan barbarians from the steppes. In fact, it was the result of the slow breakdown of centralized authoritarian power, which did not leave behind a wasteland but rather gave way to more modest, village-type settlements. There are no traces of Andronovan objects south of the BMAC, and the same is true in the Hindu Kush mountain passes that lead to India. As we have seen, there are no traces either in the Indus Valley. But since the current languages spoken in Northern India indeed belong to the Indo-European group, there is only one solution left to save the invasionist model, or at least the concept of an “arrival of the Indo-Iranians”: invisible migrations.