If you're talking about the need for empirical support for conceptual arguments, I agree that empirical evidence is always helpful and sometimes crucial. Some of my writings are more empirical or conceptual than others. It depends on my mood and on how I plan to break down the topic. Sometimes I cover other aspects in other articles. Or sometimes I'd rather just do the conceptual work because it's easier for me than doing a lot of research.

There's also a division of labour here. It's possible to focus on the big picture and to let others support the premises with empirical studies. The two kinds of writing would be complementary.