Multimodal metaphors in a Sino-British co-produced documentary

The study situates itself in the rapid expansion of multimodal communication enabled by multimedia technologies, where texts combine language, images, sound, movement, and music. Multimodal discourse research has evolved from social-semiotic analyses to integrated approaches that include corpus linguistics and cognitive metaphor theory. While multimodal metaphor research has flourished in static genres such as print ads and cartoons, dynamic video genres like documentaries remain underexplored. Given documentaries’ role in knowledge transmission and national image-building, the paper investigates how multimodal metaphors are represented in the Sino‑British co-produced documentary Through the Seasons: China, what modal characteristics they exhibit, and why they take such forms, aiming to improve understanding of metaphors in documentaries and refine theories of multimodal metaphor.

Early multimodal metaphor studies concentrated on static media (print advertising, political/editorial cartoons), examining pictorial metaphor taxonomies (e.g., Forceville’s MP1/MP2/PS/VPM; later contextual/hybrid/pictorial simile/integrated metaphors) and genre-specific configurations. Subsequent work extended to film, animation and promotional clips, addressing salience, expressive action, and spatio‑temporal metaphors. Feng (2011) proposed cross‑modal, mono‑modal, and multimodal mappings; Lan and Zuo (2016) added implicit mapping; Yu (2013) incorporated static‑dynamic and concrete‑abstract dimensions. Despite advances on meaning construction in films, the representation patterns of multimodal metaphors in documentaries are understudied. The review motivates a genre-sensitive, systematic classification of multimodal metaphor representations in dynamic discourse.

Corpus: the Sino‑British co-produced documentary Through the Seasons: China (3 episodes; ~150 minutes; released January 2022; publicly available on Youku). Research questions: (1) How do the modes coordinate to represent multimodal metaphors and what are the main representations? (2) What modal characteristics do they display? (3) Why do they present such features? Identification: Applied Forceville’s criteria to detect metaphors construable as A IS B across two or more symbol systems/sensory modalities. To reduce subjectivity, used Cameron & Maslen’s procedure; two researchers independently identified metaphors using NVivo 12 Plus; inter-coder agreement 88.28%. Of 17 discrepancies, 11 resolved by discussion; 6 discarded. Total identified and analysed: 139 multimodal metaphors. Coding: Three-stage NVivo coding. Open coding marked modes for source/target (verbal A1/B1; pictorial A2/B2; implicit A/B). Axial coding grouped representations by configurations (e.g., A1B2; A1A2B). Selective coding integrated to five subcategories: cross‑modal mapping (CRM), mono‑modal mapping (MOM), multimodal mapping (MUM), single‑domain implicit mapping (SIM), dual domain implicit mapping (DUM), and two top-level categories: explicit (EXM) vs implicit (IMM). Analyses combined quantitative frequency distributions with qualitative case studies to uncover patterns, modal relations (verbal–pictorial), and cognitive rationales.

Classification and frequencies: Two main categories and five subcategories (further split into eight models). Of 139 metaphors: implicit representations 110 (80%)—single-domain implicit mapping 105 (76%: implicit source–verbal target–pictorial supplement 42; implicit source–verbal and pictorial target 63) and dual domain implicit mapping 5 (4%: verbal and pictorial supplement). Explicit representations 29 (20%)—mono‑modal mapping 25 (18%: verbal mapping–picture example source 4; verbal mapping–picture example target 21); cross‑modal mapping 2 (1%: verbal source–pictorial target 1; pictorial source–verbal target 1); multimodal mapping 2 (1%: verbal–pictorial mapping 2). Mode usage (Table 8): Source domain—implicit 110 (79%), verbal 22, verbal‑pictorial 6, pictorial 1. Target domain—verbal‑pictorial 86 (62%), verbal 47, implicit 5, pictorial 1. Verbal–pictorial relations (Table 9): Juxtaposition 70 cases; Interpretation 67; Complementation 2. Representative examples: cross‑modal (COMMUNITY IS A TREE), mono‑modal (WETLANDS ARE SPONGES), multimodal (WILD ANIMALS ARE BABIES), single‑domain implicit (TEA TREES ARE HUMANS), dual domain implicit (GOOD IS UP). Overall, the dominant pattern is implicit source with verbal and/or pictorial target, and the most frequent relations between modes are juxtaposition and interpretation.

Findings address the research questions by showing that documentaries favor implicit mappings, particularly implicit source with verbal and pictorial target, aligning with the genre’s aims of authenticity and audience-friendly exposition. Verbal–pictorial co-deployment supports recognition and disambiguation of abstract targets, while source domains are often implicit due to their conventionality and universality. The uneven modal distribution reflects genre constraints: documentaries avoid overtly forcing metaphors, rely on visuals for concreteness, and use narration to anchor meaning. Audience embodiment and culture guide representation choices: bodily-grounded sources (e.g., sponge, human) facilitate cross-cultural comprehension; culturally specific interpretations (e.g., Chinese ‘root’ culture) are supported by verbal cues to ensure recognition. The study refines existing classifications by integrating explicit/implicit perception with source/target modal configurations and mapping relations, and argues for a broader, functionally grounded definition of multimodal metaphor beyond mere mode-differentiation of source/target. The balanced prevalence of juxtaposition and interpretation suggests a dual strategy that affords redundancy and accessibility, enhancing communicative effectiveness.

The paper proposes a unified, genre-sensitive classification of multimodal metaphor representations in documentaries, distinguishing explicit (cross‑modal, mono‑modal, multimodal) and implicit (single‑domain, dual domain) mappings, with single-domain implicit mapping as the dominant pattern. It documents modal configuration tendencies (implicit for sources; verbal‑pictorial for targets), summarizes prevalent verbal–pictorial relations (juxtaposition, interpretation), and revises the definition of multimodal metaphor to encompass source/target representational modes, mapping relations, and overall modal perception. The contributions offer theoretical refinement of multimodal metaphor, practical guidance for documentary creation (what to communicate, how, and to whom), and strategies for enhancing viewers’ multimodal literacy. Future research should expand samples across more documentaries and genres to test generalizability and further calibrate the classification.

The analysis is based on a single documentary with a limited sample size. Cultural specificity of interpretations and coder subjectivity, though mitigated by reliability checks, may affect generalizability. Future work should broaden corpora and conduct cross-cultural audience studies.