The Gold Standard of Anthropology | Deictic Reference, Demonstratives and Shared Visual Space

1. Begin with a deictic centre, then watch it move

Back to contents

Deictic expressions are interpreted relative to an origo: a practical centre involving speaker, place and moment. English contrasts this and that, here and there, now and then, but languages divide the field in many other ways. The centre is not permanently fixed to the speaker’s body. People can adopt the addressee’s viewpoint, project themselves into a map or story, or refer from the imagined position of a future user.

Record who is oriented where when the expression occurs. A claim about “near” or “far” is incomplete unless the analyst shows whose perspective, which sensory field and which activity make that relation relevant.

2. Treat demonstratives as instructions for attention

Back to contents

A demonstrative does not normally identify a referent by descriptive content alone. It tells recipients to search a contextually organised field. Gesture, gaze, head movement, body orientation, object salience and prior talk narrow the possibilities. A noun phrase such as “that red cup” combines an attention-directing term with descriptive material; “that” alone demands more from the shared scene.

The analytic unit is therefore multimodal. Align the word with gesture onset and stroke, gaze shifts, recipient orientation and the next action. A transcript that preserves only speech can make a situated reference look vague when it was locally exact.

3. Do not reduce demonstrative systems to distance

Back to contents

Near–far contrasts are useful starting points, but demonstrative choice can also reflect reachability, ownership, visibility, elevation, attention, emotional stance or whether an object lies inside a conversational dyad. Face-to-face participants may treat the space between them as shared even when an object is physically closer to one person. Side-by-side participants can organise the same room differently.

Describe the participant constellation and test rival explanations. If this is used for both near and distant objects already under joint attention, raw distance is not enough. Cross-linguistic categories should be learned from systematic use, not forced into an English two-term template.

4. Build shared visual space turn by turn

Back to contents

Shared visual access is not guaranteed by co-presence. A display may be visible to one participant, a shelf may block another, and a recipient may be looking elsewhere. Speakers recruit attention with names, gaze checks, pointing, repositioning and demonstratives. Recipients display access by shifting gaze, identifying a candidate or acting on the object.

Treat common ground as publicly managed. The analyst can show that participants behave as if an object is mutually available, then revise that claim when repair appears. Avoid saying that both people “saw the same thing” unless the record provides evidence of coordinated orientation and consequential uptake.

5. Distinguish locating, selecting and contrasting

Back to contents

“This” can introduce an object, select it from competitors, contrast it with another, claim present relevance or organise discourse. A pointing gesture may identify a region rather than a single item. Repeated demonstrations can divide a workspace into here versus there or establish named zones that remain available after the gesture ends.

Code the action accomplished, not just the lexical form. Ask what alternatives were live and what the recipient did next. If a cook says “put these there” while touching two bowls and indicating a counter, the plural touch, direction of point and subsequent placement jointly resolve two different deictic tasks.

6. Worked example: three similar objects in a museum case

Back to contents

In an invented museum discussion, a guide says, “This was repaired later,” while pointing through glass at three bronze pins. The visitor looks to the centre pin and asks, “The one with the bright tip?” The guide shifts her hand left and replies, “No, this one—the bent pin beside it.” The visitor nods and photographs the left pin.

The first token did not specify a stable geometric distance; it initiated selection inside a crowded visual field. The candidate description exposed the visitor’s interpretation, and the guide repaired the gesture and added contrastive description. Successful photography supplies stronger evidence of resolution than the nod alone.

7. Use repair to reveal the search field

Back to contents

Repair initiations such as “Which one?”, a candidate point or a puzzled gaze reveal that the deictic instruction did not yet isolate a referent. Speakers may hold a point longer, move closer, add a noun, contrast alternatives or reposition the recipient. Those changes show what information the first design lacked.

Preserve the entire trajectory from initial reference to next action. Do not label every delay as misunderstanding: a recipient may be physically locating an item. Conversely, immediate compliance may still select the wrong referent. Later correction, object handling or task failure can reveal a mismatch hidden by smooth talk.

8. Changed-context transfer check: remote video repair

Back to contents

Move from co-present museum talk to a technician guiding a remote user through a video call. “Press that button” depends on camera framing, screen mirroring, transmission delay and whether pointing occurs in the technician’s space or on a shared annotation layer. The user may see left and right reversed.

The transferable method is to map each participant’s visual field and the transformation between them. A screen cursor, verbal landmark and numbered label can replace an embodied point. Shared visual space is now technologically produced, so the audit must include interface state, lag and what each side could actually see.

9. Know the limits and common misconceptions

Back to contents

Deixis is not a universal code in which this always means near and that always means far. Demonstratives can be ironic, affective, discourse-organising or projected into imagined space. Pointing direction does not by itself establish the target; hand shape, trajectory, gaze and activity matter. Cultural conventions also shape acceptable pointing forms.

Experimental tasks can isolate distance while narrowing ordinary social uses. Ethnographic recordings preserve context but may offer fewer controlled contrasts. Generalisations should survive both natural collections and structured comparisons, with language-specific categories reported on their own terms.

10. Audit the multimodal reference chain

Back to contents

Retain synchronised video from angles that show hands, faces and candidate referents. Document seating, object layout, visibility barriers and changes in the scene. Transcribe demonstratives, nouns, prosody, gaze, gesture phases, recipient movements, repair and the action that follows. State how cases entered the collection.

For each instance, identify the proposed deictic centre, search field, competing candidates and evidence of resolution. Compare distance, shared-space, accessibility and discourse explanations. Publish ambiguous and failed cases so readers can test whether the preferred account explains more than a polished success sequence.

11. Frequently asked questions

Back to contents

Is this always nearer than that? No; distance is only one possible organising factor. Does pointing make a reference unambiguous? Not when several candidates fall along the same vector or recipients lack visual access. Is shared visual space the same as common ground? It is one publicly managed resource, not proof of identical knowledge.

Can deixis occur without visible objects? Yes, in discourse, time, maps, narratives and imagined space. Does a nod prove recognition? Not necessarily; subsequent action is stronger evidence. Must every language have English-like demonstratives? No; systems vary in form and contrast.

12. The gold standard keeps the scene inside the analysis

Back to contents

Deictic reference works because participants coordinate language with bodies, perspectives and material surroundings. Demonstratives do not merely label distance. They organise attention inside a scene whose boundaries can be negotiated, repaired, projected or technologically transformed.

The gold standard therefore connects each form to the visible field, participant constellation, competing candidates and consequential uptake. It keeps failures and alternative explanations available. That turns small words such as this and there into precise evidence about how people build a shared world for the next action.