Why does silicon rich, and therefore non-viscous lava occur at destructive boundaries and vice-versa?

I'm writing an essay comparing the Sufriere Hills volcano in Montserrat and Mt. Nyiragongo near Goma in the DRC. I've learnt that although they are both stratovolcanoes/composite volcanoes, the lava at Montserrat (basalt) is much more rich in silica than the highly alkaline, rhyolite lava in Goma, so that eruptions in the former are explosive whilst in the latter they are more like a fast 'trickle'. How can I explain this in the context that Montserrat is the product of the North American-Caribbean destructive boundary, whilst Nyiragongo formed on the horst on the East African Rift valley as a result of a constructive fault line?