In their 2016 work, Aguirre-Urreta et al. provided a contribution to the literature on causal measurement models that enhances clarity and stimulates further thinking. Aguirre-Urreta et al. presented a form of statistical identity involving mapping onto the portion of the parameter space involving the nomological net, relationships between the variable and other variables external to the scale (T[subscript n]). The argument does not require that the same variable cannot have different nomological nets in different contexts but does require that two variables cannot share the same nomological net. The conclusion requires a sufficiently elaborate nomological net to render this one-to-many mapping plausible. The central idea involves showing that indicators do not determine the identity of the latent variable because one can omit indicators without changing the effects of the latent variable. Enthusiasts of causal indicators might point to the distinction Bainter and Bollen (2014) made between data-driven approaches that they booed and theory-driven approaches that they cheered. Enthusiasts might argue that Aguirre-Urreta et al.'s focus on empirical constraints on variable identity relied on a data-driven approach. By asserting that the researcher determines the identity of the latent variable, the claim denies that anything in the model performs this task. Even from a theory-driven perspective, however, the empirical constraints on variable identity make the theory testable. So, both approaches rely on such constraints. The message for enthusiasts of causal measurement seems clear: Stop defending and start extending. The existing body of theory remains insufficiently developed to defend causal measurement models and the debate can only move forward on the basis of more specific and elaborate theories. The corresponding caution for skeptics is that while contributions such as that of Aguirre-Urreta et al. can help prod enthusiasts toward greater clarity, it remains too early in the game to draw firm conclusions about causal measurement models until enthusiasts can provide a clear and reasonably complete account.