This paper argues for a fundamental reorientation in how we understand and analyze pragmatic action in human interaction. While the multimodal turn in pragmatics and conversation analysis has successfully challenged logocentric assumptions, a persistent ontological separation of modes continues to underpin much of contemporary research. We propose integrating a translanguaging lens into Multimodal Pragmatics and Interaction Analysis, drawing on the insight that just as named languages are ideological constructs imposed upon a continuous linguistic flow, named modes are analytical constructs imposed upon a continuous semiotic flow. Through three detailed illustrative examples drawn from video-recorded naturalistic interaction in educational, workplace, and medical settings, we demonstrate that meaning emerges from fused semiotic gestalts rather than from the coordination of independent streams. Our analysis reveals how participants orchestrate unitary semiotic repertoires in real time, how implicature operates through multimodal tension, and how institutional power operates through semiotic hierarchies. We conclude by discussing the theoretical implications for Gricean pragmatics, communicative competence, and the concept of context, while acknowledging methodological challenges around transcription and reproducibility. This framework offers a more ecologically valid and ethically responsible approach to understanding human communication.