J2 is nearly absent in northern Europe and western Europe. The distribution of J2 in much of Europe also mirrors the expansion of the Roman Empire, from which I conclude that J2 was probably restricted to the Mediterranean (Italy and southern Balkans) before the classical period. Which excludes it from being associated with the spread of Indo-European languages.
Well, if you think about it, there should be no I2*. Bear in mind that all individual subclades of Haplogroup I (or, I should say, Haplogroup I1, and all subclades of Haplogroup I2) are essentially Neolithic in age. That is, they represent individual male lineages that survived the Neolithic. For instance, Haplogroup I2-M26 is thought to be ca. 8000 years old.
I* (ancestor of I1 and I2) is far older than PIE in any case - it probably appeared around or slightly before the last glacial max.
That's also impossible, due to the same reason (the age/timeframe).
In my opinion, R1a-M417 is associated with the Kurgan culture. Asian R1a is mostly R1a-Z93, which is a subclade of R1a-M417. In contrast, European R1a is dominated by the other main subclade of R1a, R1a-Z283. Overall, I find this quite consistent with the Kurgan scenario.