ABSTRACT Investigations of some Type Iax supernovae (SNe Iax) have led to the suggestion that their ejecta must be layered to some degree. Such an ejecta structure has been argued as inconsistent with the well-mixed composition predicted by pure deflagrations. Based on explosion models, we create toy models in which the ejecta are artificially stratified and progressively mixed until a uniform composition is obtained. We find that models that are heavily mixed, containing burned and unburned material at all velocities, produce reasonably good agreement with SN 2012Z, for which a layered structure has been suggested. We also discuss how existing ejecta compositions determined for SNe Iax do not necessarily contradict pure deflagration models and may be consistent with a steeper density profile. We investigate previous claims that differences in line profile shapes may be due to strong blending, by presenting a series of models with different plasma states. These models indicate that blending could indeed explain differences in the observed profiles. Alternatively, stratification could also explain such differences; however, all of our models indicate that this does not necessarily require stratification in abundance. Sufficient stratification in the ionization state can be achieved even for a well-mixed model. Based on our analysis, we demonstrate that there is insufficient evidence to suggest the ejecta of SNe Iax must be layered and therefore argue the pure deflagration scenario is not ruled out, even for the brightest SNe Iax. Our analysis does not indicate the ejecta cannot be layered to some degree, but observations within days of explosion are necessary to determine the extent to which the outer ejecta could be layered.