Cinematic virtual reality (CVR) is an audio-visual form viewed in a virtual reality headset. Its novelty lies in the way it immerses its audience in highly realistic 360° visual representations. Being camera-based, CVR facilitates many of the practices of conventional filmmaking but fundamentally alters them through its lack of a rectangular frame. As such, CVR has garnered scholarly attention as a ‘frameless’ storytelling medium yet to develop its own language. The form has gained traction with producers of nonfiction who recognize CVR’s capacity to transport audiences to remote social worlds, leading to claims that equate CVR’s immersion with a social and emotional response to its filmed subjects. A strand of CVR scholarship has emerged, grounding nonfiction CVR theoretically and critiquing such deterministic claims. Broadly speaking, these parallel strands of inquiry point to a common concern with CVR’s semiotics; as the meaning potential of the 360° format, and the social aspects of its use in documenting reality. Currently however, there appears to be a lack of systematic analyses that foreground CVR’s semiotics. This study addresses this gap by using social semiotic methods to complement these threads of inquiry, subsuming them into a holistic account of CVR’s semantics. Utilizing systemic functional methods, multimodal discourse analyses were performed on nonfiction CVR texts addressing core research objectives. The first objective is the systematic description of CVR as a semiotic technology, and the configuring of discourse through its novel 360° modality. The CVR spectator is described for their role in the real-time construction of low-level meanings. Higher-level concepts further characterize CVR texts as technologically enabled, virtual sites of social discourse. The second research objective concerns clarifying the implications of CVR for nonfiction practitioners. Nonfiction discourse is conceptualized as the negotiation of semiotic autonomy, independence, and control, between viewing spectator, filmed subject, and CVR author respectively. The third objective concerns the development of an analytical approach tailored specifically for CVR. Extant systems from image, text, film, and action analyses are reflexively applied, appraised, and adapted for use in the study of CVR and new frames are presented to cater for the 360° modality. The findings show CVR to be an inherently logical, contextualizing form, where the spectator has a degree of sense-making autonomy in the construction of representational and social meanings. This semantic autonomy is found to camouflage the deeper textual constructions in what appear as ‘reality experiences’. The repercussions for the CVR producer are the indeterminacy of meanings which are ‘at risk’ in particular ways when conventional framing methods cannot be utilized, and when the spectator is given reflexive agency to make meaningful connections across the 360° image. Systemic functional analytical methods prove flexible enough to be applied to the texts, and open enough for the study to present additional systems and frames for a more fulsome approach to the analysis of CVR.