[DSRP Evidence](https://dsrpevidence.org/)

# Can language restructure cognition? The case for space

## Details

**Authors** Majid, Asifa and Bowerman, Melissa and Kita, Sotaro and Haun, Daniel B.M. and Levinson, Stephen C.

**Year** 2004

**Publisher** Trends in Cognitive Sciences

**Kind of work** article

**Discipline** Linguistics

**Secondary disciplines** Cognitive Science, Anthropology

**Applied** false

[Read it at the publisher](https://doi.org/10.1016/j.tics.2004.01.003) 
10.1016/j.tics.2004.01.003

## In authors' words

### Abstract

Frames of reference are coordinate systems used to compute and specify the location of objects with respect to other objects. These have long been thought of as innate concepts, built into our neurocognition. However, recent work shows that the use of such frames in language, cognition and gesture varies cross-culturally, and that children can acquire different systems with comparable ease. We argue that language can play a significant role in structuring, or restructuring, a domain as fundamental as spatial cognition. This suggests we need to rethink the relation between the neurocognitive underpinnings of spatial cognition and the concepts we use in everyday thinking, and, more generally, to work out how to account for cross-cultural cognitive diversity in core cognitive domains.

### What they set out to do (purpose)

To review cross-linguistic and cross-cultural evidence on spatial frames of reference and argue that language can structure or restructure spatial cognition.

### Who or what was studied (sample)

Review of studies of speakers of about 20 languages; nonlinguistic rotation-task results reported for speakers of ten communities, with detailed results for Dutch and Tzeltal; gesture studies; child acquisition data for English, Italian, Serbo-Croatian, Turkish, Tzeltal, Tzotzil and Balinese.

### How they did it (methods)

Review (opinion article) combining linguistic description of frames of reference, nonlinguistic tasks in which participants view an array, are rotated 180 degrees, and then reproduce or reason about it, analysis of spontaneous co-speech gesture, and developmental data. Ecological zone, dwelling type, subsistence mode and individualism/collectivism are tabulated against each language's preferred frame.

### What they found (results)

Languages differ in which of three frames of reference (Relative, Intrinsic, Absolute) they make available and prefer for table-top space; Guugu Yimithirr uses only the Absolute frame. On nonlinguistic rotation tasks testing memory for configuration, memory for path and transitive inference, speakers' solutions align with their language's preferred frame: Dutch responses were overwhelmingly Relative and Tzeltal responses predominantly Absolute, with over 80% of Tzeltal speakers coding consistently in Absolute terms on one task. Spontaneous gestures follow the language's frame. Children acquire Absolute systems as early as, or earlier than, Relative ones. No ecological, subsistence or cultural-style variable determined the frame, though urban dwelling was associated with the Relative frame. The authors conclude that language can play a central role in restructuring cognition.

## Commentary

### In short

People describe where things are using different reference systems: left of, north of, or at the front of an object. This review finds that the system a language favors is also the one its speakers use when solving spatial puzzles without words.

### In more detail

The authors distinguish three frames of reference. The Relative frame locates one object with respect to another from a viewer's standpoint (the fork is left of the spoon); the Intrinsic frame uses the parts of the ground object itself (the fork is at the nose of the spoon); the Absolute frame uses fixed bearings such as north or uphill. The Intrinsic frame requires parsing objects into facets such as front and side, and the Relative frame requires parsing the self into front, back, left and right and projecting that onto the scene, encoding a three-way relation among object, landmark and viewer; a two-way relation between self and object remains Intrinsic. The Relative frame can also be transferred to another person's viewpoint. Because the Absolute frame stays fixed when a person turns around while the Relative frame rotates with them, rotation tasks separate the two. Across such tasks, speakers of Absolute languages (including Guugu Yimithirr, Arrernte, Hai//om, Tzeltal, Longgu, Belhare and Balinese) solved problems in absolute terms and speakers of Relative languages (Dutch, Japanese, Yukatek) in relative terms, and spontaneous gestures showed the same split. Tzeltal children used absolute vocabulary by age two and Balinese four-year-olds behaved absolutely on nonverbal tasks, while English-speaking children use left and right for object-to-object relations only around age 11 or 12. A table of 20 languages showed no simple ecological or cultural predictor of frame. The authors propose perceptual tuning, re-representation, structure-mapping and computational cost as mechanisms by which language could shape nonverbal coding.

### Response

Frames of reference are a direct case of P: the same table-top scene receives different, systematically related descriptions depending on whether the viewer, the landmark object or a fixed external bearing supplies the point of view, and the viewer's frame can be lifted and applied to another person. The article adds an effect claim, that the perspective a language makes habitual shapes nonverbal reasoning, memory and gesture. Its account of how the frames work also requires R, since the Relative frame encodes a three-place relation among figure, ground and viewer that the authors distinguish from a two-place one, and S, since both the Intrinsic and Relative frames depend on parsing objects and the self into parts. Caveats: the evidence for the effect is correlational across populations and the causal direction has been disputed (Li and Gleitman 2002); the article is a review that draws mainly on its authors' own studies; and its stress on cross-cultural variation fits DSRP's universality claim only if what is universal is the capacity to adopt a frame, not any particular frame.

**Patterns it shows** S, R, P

Shows the structure is there, Shows it makes a difference

**Added** 2026-11-19

**How to cite this** Majid, Asifa and Bowerman, Melissa and Kita, Sotaro and Haun, Daniel B.M. and Levinson, Stephen C. (2004). Can language restructure cognition? The case for space. Trends in Cognitive Sciences.
