David Morgan's analysis of the concept of triangulation (Morgan, 2019) raises some important issues for mixed methods researchers. I agree with Morgan's general critique of this concept, and his account of the origin of the term's use in social science provides a valuable clarification of its history. 1 However, neither Morgan's analysis, nor the earlier discussion of this concept by Fetters and Molina-Azorin (2017), makes the important point that "triangulation", aside from its original use in surveying, is a metaphor. Like all metaphors, triangulation provides some insights into the phenomena to which it's applied, while ignoring or obscuring other insights (Lakoff & Johnson, 1980, 2003. This in itself is sufficient reason to abandon triangulation as a technical term in the social sciences. 2 In particular, the "triangulation" metaphor obscures the fact that qualitative and quantitative methods provide very different kinds of evidence about the phenomena studied. Morgan states that "Complementarity assigns different goals to the qualitative and quantitative portions of a project, according to the strengths each method offers for a particular purpose", but he never discusses what these different strengths are. (I think that this is also true of many mixed methods textbooks).The main strength of quantitative methods is in defining different variables, measuring these, and showing relationships among variables; experimental methods can also provide strong evidence that these relationships are causal, although there are important limitations on this; see Maxwell (2016Maxwell ( , 2021. The main strength of qualitative methods, on the other hand, is in understanding the processes (including mental and cultural processes) through which some events or situations influence other events or situations-showing how these relationships occur. Shadish, Cook, and Campbell, in their book Experimental and Quasi-experimental Designs for Generalized Causal Inference (2002), argued that the unique strength of experimentation is in describing the consequences attributable to deliberately varying a treatment. We call this causal description. In contrast, experiments do less well in clarifying the mechanisms through which and the conditions under which that causal relationship holds-what we call causal explanation (p. 9; emphasis in original).They later discuss how qualitative methods can discover and verify causal explanations (pp. 389, 391). For a more detailed presentation of the relative strengths and limitations of qualitative and quantitative approaches to causation, see Maxwell (2020Maxwell ( , 2021.This complementarity of strengths and limitations has major implications for combining the two methods, and challenges Morgan's (and others') division of possible outcomes into