Space in Language and Cognition

Content

1. Introduction

2. Spatial reasoning and language
2.1 Frames of reference: linguistic diversity in spatial coding
2.2 Linguistic relativity: Pederson’s findings and interpretations
2.3 Universal findings: Li and Gleitman’s contraposition
2.4 Defending linguistic relativity: Levinson’s answer

3. Discussion

4. Literature

1. Introduction

The relation between language and cognition has already occupied philosophers as John Locke, Johann Herder and Wilhelm von Humboldt centuries ago. In the early 20th century, the linguists Franz Boas, Edward Sapir and Benjamin Lee Whorf contributed to making it a big topic in the scientific world. Even today, the connection between the language people speak and the way they think and view the world remains a major contemporary research area in linguistics, cognitive psychology and anthropology. The main questions are: Do languages have an impact on the way their speakers think? Or is it the other way round: Does cognition channel language? Which comes first: language or thought?

The problem here will be approached by focusing on the relation of spatial coding in language and cognition. Space constitutes a central domain for humans with directions and locations being of crucial importance in our everyday life. Findings in this domain might be subject to generalizations on other, less central domains of the human experience.

This term paper deals with the debate between two opposing viewpoints about the causation between language and cognition in the spatial domain. The first perspective is called linguistic relativity, which holds that the structure of a language influences the cognitive processes of the speakers and affects the ways in which they conceive the world. In other words, the principle of linguistic relativity claims that language shapes the way we think. This standpoint will be represented by articles from research groups around the linguists Eric Pederson and Stephen Levinson (Pederson et al., 1998; Levinson et al., 2002)^[1]. The opposing stance contains the universalist notion that all languages are broadly similar and linguistic systems are merely the formal and expressive medium that speakers use to describe their mental representations. Hence, linguistic coding cannot have effects on cognition but reflects antecedently existing conceptual distinctions. This attitude is conveyed by the psychologists Peggy Li and Lila Gleitman (Li & Gleitman, 2002), who argue that “it’s the thought that counts” (ibid, 291).

The paper is structured in the following way. To explain the basis of the debate, the main part opens with an introduction of the different frames of reference that are employed in distinct languages to convey spatial information. Next, Pederson’s experiments and interpretations about the effect of spatial coding in language on cognition are presented. Then, a reinterpretation of these findings and further experiments by Li and Gleitman are outlined. Finally, a critical comment by Levinson on Li and Gleitman’s analysis will show how he defends the position of linguistic relativity in the domain of spatial coding. In the concluding chapter of this paper, a summary of the debate will be provided, accompanied by a discussion about the transferability on the general relation between language and cognition.

2. Spatial reasoning and language

2.1 Frames of reference: linguistic diversity in spatial coding

It may seem ironic that Benjamin Lee Whorf, the forefather of linguistic relativity, stated that the realm of space constitutes an exceptional domain in that nonlinguistic representations of space are independent of the language one speaks. Whorf says: “But what about our concept of ‘space’ […]? There is no such striking difference between Hopi and SAE about space as about time, and probably the apprehension of space is given in substantially the same form by experience irrespective of language” (Whorf, 1956, 158). In the 1990s, this universalist claim has been seriously challenged by a group of researchers from the Radboud University and the Max-Planck Institute in Nijmegen, Netherlands (Haviland 1993; Pederson et al., 1998, among others). The most interesting finding negates Whorf’s first statement that there is ‘no such striking difference’ in spatial reference between languages. The Dutch researchers found that distinct languages do differ in the way they can be used to describe spatial matters by employing different frames of reference.

Pederson developed the men-and-tree-game to demonstrate this evidence. The game consists of two people from the same community playing director and matcher. The two are seated at a table next to each other, but they are not able to see each other due to a screen between them, as shown in figure 1 (Pederson et al., 1998). They each have the same set of photographs in a different order on the table. The task is for the director to describe his pictures one by one. The matcher is supposed to match his pictures accordingly and may ask questions in between. Among the pictures is a target subset of photographs depicting a toy man and a toy tree. As seen in figure 2 (Pederson et al., 1998), the arrangement of these figures differs in orientation to each other. Also, the toy man is facing in different directions from the viewer’s perspective. The interactive discourses during the men-and-tree-game have been recorded, transcribed and translated. By examining the spatial descriptions of the pictures, Pederson could draw a typology to characterize the linguistic preferences of speakers from different communities, such as Dutch, Japanese, but also Australian, African and Mayan languages.

illustration not visible in this excerpt

Figure 1: Arrangement for playing the men-and-tree-game

illustration not visible in this excerpt

Figure 2: Photographs from the target subset in the men-and-tree-game

Altogether, Pederson found three distinct spatial frames of reference in the language data. The expression ‘frame of reference’ in this context can be defined in terms of the Gestalt psychology as “a unit or organization of units that collectively serve to define a coordinate system with respect to which certain properties of objects, including the phenomenal self, are gauged” (Rock, 1992, 404). Pederson differs between three such frames of reference in languages, which he calls ‘relative’, ‘intrinsic’ and ‘absolute’. The relative frame of reference uses viewer-centered coordinates based on body axes (left/ right/ front/ back) to describe the location of objects. The second picture in figure 2 (2.2) for example could be described as ‘The tree is right of the man’ (that is, right from the viewer’s perspective). The intrinsic frame of reference employs object-centered coordinates of the reference or landmark object based on ‘intrinsic’ facets of the object, as in the sentence ‘The tree is at the back of the man’. In languages with the absolute frame of reference, objects are located in terms of coordinates based on fixed bearings such as cardinal directions, as in ‘The tree is north of the man’.

The men-and-tree-game showed that there are languages (e.g. Arrernte in Australia, Longgu in Austronesia and Tzeltal in Mexico), which make use of the absolute frame of reference and which lack all terms that are relative to body orientation as ‘right’ or ‘left’. A sentence as ‘The tree is left of the man’ is simply not translatable into those languages. Pederson notes that a specific language can have the means to articulate either one, two or all three frames of reference.

[...]

^[1] For a better readability the two research groups will from now on be referred to as ‘Pederson’ and ‘Levinson’.