Unnatural narrative theory, initiated by Brian Richardson in his book Unnatural Voices (2006), has had an invigorating effect on narratological debates on both sides of the Atlantic. The unnatural is an intriguing concept that offers a fresh perspective on established narratologies, especially those based on a cognitive framework. However, as the critical responses to Richardson's keynote lecture at the ENN Conference in Paris and to the Unnatural Narrative Panel at the ISSN Conference in Manchester (both in 2013) show, many narrative theorists remain skeptical of the new approach. It therefore seems timely to devote a special issue of Style to Richardson's contribution to narrative theory. In a now well-rehearsed manner, the Target Essay seeks to establish the relevance of this new approach by arguing not only that a large body of texts has hitherto been ignored by narratologists, but also that most existing narrative theories are conceptually flawed. How exactly unnatural narrative theory intends to set things right is only mentioned in passing: two pages refer the reader to other work both by Richardson and others. In short, we learn less about unnatural narrative theory itself than about why it is needed. Instead of engaging with this rhetoric of innovation, my response will focus on the premises and methodological foundations of unnatural narrative theory. I will briefly outline how narratological theory design usually works, starting with definitions of narrative that allow us to deduce further hypotheses, to develop models, and to propose terminologies in order to produce systematic knowledge about the forms, functions, and uses of narrative (note that terms like deduction, induction, inferences, abstraction, and knowledge are used more loosely in literary theory than in formal logic and epistemology). For reasons of space, my short introduction will focus on the kind of reasoning typical of text-oriented rather than cognitive approaches, as Richardson expresses a clear preference for the former; this does not mean, however, that I share his view of cognitive narratology in general or of David Herman's pioneering work in particular. In the body of the article, I will then juxtapose my account of narratological theory design with Richardson's inductive approach in order to discuss whether unnatural narrative theory fails to propose a coherent theoretical argument (one of the objections often raised by critics of the unnatural approach). It should be noted that narratology, as the systematic study of narrative representations, always provides nominal, not real definitions. These are not aimed at capturing the distinctiveness of specific (sets of) narratives but mainly serve to distinguish narrative from nonnarrative and antinarrative (cf. Prince, "On a Postcolonial" 373). The term narrative thus signifies an abstraction, an idea of what constitutes a narrative. Narrative, as an abstract entity, has been conceptualized in various ways, for instance as a text type or discourse type, a mode of representation, a form of communication, or, if viewed from a cognitive angle, a way of sense making or world making. Proceeding from such general premises, narratological approaches have developed well-constrained descriptions of narrative as a specific text type, as a verbal, auditory, or visual representation, a fictional or nonfictional form of communication; and they have proposed heuristic frameworks for theorizing cognitive parameters in order to further our understanding of how we make sense of narrative. This is not to say that narratological reasoning is wholly deductive: in practice, aesthetic experiences and individual preferences influence conceptual and terminological decisions (this is especially true for approaches that limit themselves to the analysis of narrative fiction). However, narratology has always sought to reconcile intuition and abstraction in a systematic manner (cf. Sommer). When outlining the project that eventually evolved into unnatural narrative theory, Richardson proceeded from rather different assumptions, as he explains in Unnatural Voices: This is an empirical study that describes and theorizes the actual practices of significant authors, instead of building on a priori linguistic or rhetorical categories; such an inductive approach is essential because many extreme forms of narration seem to have been invented precisely to transgress fundamental linguistic and rhetorical categories. …