In a series of recent papers, we have suggested that debates in a large number of areas—including topics of significant interest to philosophers, and even debates within philosophy—can best be characterized as what we call metalinguistic negotiations. Speakers involved in a metalinguistic negotiation do not assert conflicting contents. Rather, they negotiate how to use their words, which concepts to employ, how to set contextual parameters, how to precisify vague terms, and similar representational matters. This type of analysis is tailor-made for domains like aesthetic and taste discourse, and its plausibility seriously undermines central arguments for competing, assessment-relative theories in those areas. But the picture can extend well beyond domains that are characterized by "faultless disagreement". Our choices about which concepts to use, and about which concepts to pair with which words, can be loaded with significance, in ways that correspond closely to the object-level matters that intuitively are at issue in the domains we discuss. The normative significance of our representational choices, and the correspondence between those choices and our non-representational activities and goals, ensures that metalinguistic negotiation can serve as a means for debating issues of great practical and even theoretical significance. In this talk, we introduce our framework for thinking about this mode of communication, and we gesture towards some arguments that it may be very much more common than one might think.