Soil biodiversity has become a major area of research over the last decade, and the literature on the topic has expanded tremendously in recent years, so much so that a huge number of publications now deal with soil biodiversity every year. This article does not attempt the formidable task of drawing a general picture of where the field is at the moment, but it zeroes in instead on two perspectives that seem to have gathered momentum over time and raise concern about future progress. The first perspective involves the implicit assumption that to make sense of either the species-, genetic-, or functional biodiversity of soils, it is not necessary to consider in detail the features of (micro)habitats provided by soils to organisms, and that analysis of the information provided by extracted DNA or RNA suffices. The second perspective is associated with research on the effect of the physical and chemical characteristics of microhabitats on the activity of microorganisms. It basically hypothesizes that all microorganisms behave similarly, and therefore that observations made mostly with bacteria can be extended readily to all organisms, ignoring taxonomic biodiversity. To illustrate both perspectives, we provide a number of illustrative examples from the relevant literature and analyze them briefly. We argue that these two perspectives, if they spread, will hinder progress in our understanding of soil biodiversity at any level, and especially of its impact on soil processes. In order to return to a more fruitful middle ground, where both a variety of organisms and the characteristics of the microhabitats where they reside are carefully considered, several routes can be envisaged, but our experience suggests that an emphasis on genuinely interdisciplinary research is crucial.