I cannot help being underwhelmed by papers on the so-called 'microbiome.' Very seldom is it something more complex than proportion/prevalence of species X correlates with behavior/health outcome Y. Simple correlations with all sorts of imaginable confounding variables that preclude causal explanations. Even in totally hypothesis-free, exploratory studies, almost all variation in species density over time gets explained by two or three PCA components, and gut bacterial populations seem to grow/shrink independently with respect to other populations. These dynamics, whatever they may be, are a far cry from such a lofty, puffed-up term like micro-'biome', where its use always signifies complex (and conveniently uncharacterized!) dynamics of bacterial species interacting with each other and the host.

That said, I don't find the author's framing of this particularly useful. If a theory is wrong, it will be wrong for factual, technical reasons. The extent to which some real phenomena can predict others is testable. Handwaving it away as social contagion, ideology, or mass hysteria --- not that those don't exist, medieval dance fever being my favorite example --- isn't good enough for me.