Space - Language as Spotlight (FoR) Flashcards
Levinson (2003) - FoRs
FoRs - utilised coordinate system by which spatial relationship between figure and ground can be abstracted to for communicative purposes > used when not coincident in space, otherwise topological relations would be used
Intrinsic - referring to two entities placed in respect to one another e.g., keyboard is in front of the monitor
Relative - assumes external viewpoint that imposes coordinate system on relationship between two entities with respect to one another e.g., keyboard to the right of the monitor
Absolute - uses external, objective coordinate system (cardinal directions)
Majid et al (2004) - variation in FoR use
English speakers use two different FoRs to describe spatial relationships in table-top space: relative or intrinsic, but restrict absolute to large-scale, geographical descriptions
Indo-European-family languages typically use relative FoR for table-top space, but many aboriginal languages in Australia use absolute e.g., Guugu Yimithirr doesn’t use absolute north, but similar construction (two different axes) in referring to spatial relationship between figure and ground (seen in picture description task; Levinson, 2003)
Tzeltal speakers in Mexico use absolute FoR, but uphill-downhill axis and across-axis, indicating salience of mountainous environment
Palmer (2015) - Topographic Correspondence Hypothesis
Suggests role of salient aspects of environment in cross-cultural variations in preferred FoR use
Supported by: (1) some salient aspects of environment appear to corelate with detail of systems involving absolute FoR e.g., Tzeltal; (2) Majid et al (2004) showed evidence that languages in urban environments more likely to have relative
Refuted by: (1) Majid et al (2004) no evidence that environment or subsistence type predicted FoR; (2) cannot account for why both mountainous and non-mountainous environment have absolute FoR (Tzeltal and Hai//om), and why Jahai live in mountainous region but use intrinsic
Weak support, but those that claim Majid et al (2004) only focused on assessing correlations with global factors such as ecological zone and subsistence type, rather than factors directly relevant to FoRs such as topography, mobility, and literacy (Pinker, 2007), as well as interactions with environment and impact on culture (Palmer, 2017)
Cultural Practices Hypothesis
Some observational evidence: (1) Shapero (2017) absolute FoR more common in Quechua speakers who worked as herders in highlands and travel over large distances; (2) Le Guen (2011) found men use absolute FoR more in Yucatec speaking communities, who happen to be gardeners more often
However, no quantitative study
Levinson (2003) - Navigation
Humans, in comparison to other animals like birds, relatively poor at navigation, measured in accuracy in pointing to direction of one’s start of journey when taken to novel location
At distance of only 1-2km from base, both English and Dutch (relative) poor performance, with more consistency in English pps’ answers only due to vastly more pps
Guugu Yimithirr (absolute), even at 80km from home, in novel location, excellent performance, with qualitative reports of pps using mental compass and distance estimations (cognitive computation required for maintaining effective absolute FoR; Majid et al.,2004) > roughly 10 pps, each at 20 locations, with each location normalised to North
Cross-cultural evidence key for linguistic relativity (Majid, 2018)
Criticisms of Levinson (2003) - Navigation (Pavlenko, 2014)
Pavlenko (2014) highlights pps bilingual (L1 Guugu Yimithirr, L2 English), and didn’t behave like typical English (not relative) not Guugu Yimithirr monolinguals > suggests convergence of conceptual representations of space between L1 and L2, rather than specific effect of relative vs absolute FoRs > support from same population no longer recognising cardinal directions synonymous with Levinson (2003) due to shifting towards L2 English > but no experimental study investigating this
Haviland (1980) and Levinson (1982) - Gesture
In communication, often use co-speech gesture, conveying helpful spatial information through iconic gestures and pointing (Majid et al., 2004)
Studied same Guugu Yimithirr speakers two years apart, tasked with telling same story, of a video of a boat capsizing, and in each video, orientation of North and West different directions, but gestures changed orientation, updating to real-time cardinal direction
Marghetis et al (2020) - Gesture
Delineates between culture and language
Investigated speakers of allocentric language, and allocentric-egocentric speakers, all in same village, in co-speech gesture elicitation tasks
Spontaneous gestures reliably reflected linguistic FoR, and that egocentric FoR not predicted by bilinguals’ dominant language or language used in task (not online use of language), but mastery of egocentric terms right and left
Pyers et al (2010) - Gesture
Ideal study population is cohort, with same culture and environment, but different degrees of linguistic input > same Nicaraguan Signers population as Flaherty and Senghas (2011) with number, showing cross-generational differences in consistency of spatial language, with second-cohort (younger) more consistent than first-cohort, using relative FoR and left-right language, similar to Marghetis et al (2020), and outperformed them on non-linguistic spatial cognition task
Spatial Memory
Participant faced with task, often reconstruction or spatial search, then rotated 180 degrees, forcing them to choose whether to use own body as FoR (relative; egocentric) or maintain absolute > both valid, but reconstruction memory differs on how languages encode spatial relations (Majid et al., 2004)
Majid et al (2004) experimental evidence: Dutch and Tzeltal speakers reconstructed path of moving toy after rotation, shown Dutch almost always used relative, and Tzeltal almost always absolute, with similar findings in a spatial search paradigm
Haun and Rapold (2009) observational evidence: Hai//oi-speaking child learning to dance without spatial language as instructions, validated (performed by self), then rotated, shown to maintain absolute FoR whereas German children maintained relative by swapping direction of dance moves > even own body movements, an inherently egocentric domain, encoded differently based on preferred FoR use
Spatial Memory Criticisms
Li and Gleitman (2002) argue differences result of confounds e.g., testing conditions not sufficiently controlled across populations in Majid et al (2004) > however when environmental features controlled (no available reference to topography or environment), same correlation between linguistic and non-linguistic representations found
Major variation in cultural and environmental factors within groups as large as variation between groups (i.e., within- and between relative versus absolute; Majid et al., 2004) > cannot conclusively say it is language without accounting for culture
Palmer (2017) - Sociotopographical Model
Although seems to be language driving differences, particularly in studies of co-speech gesture, proposal that differences in FoR use and consequences in non-linguistic spatial cognition tasks driven not by language primarily, but an interplay of environmental features and its affordances (to survival), topography, cultural practices, interaction with environment, and linguistic resources
Model claims it accounts for findings of same environment, different FoR, and different environment, but same FoR (Majid et al., 2004), as could be down to complex interplay of factors
Example is that of Laanu fishing and non-fishing communities, with same place, culture, and language, but occupation and subsistence accounts for allocentric vs egocentric FoR, respectively > suggests a need for more nuanced investigations
However, no causality established
Criticisms for FoR as example of language acting as spotlight - Li et al (2005)
Showed Tzeltal speakers can use non-dominant FoR, egocentric, without apparent difficulty > perhaps in previous experiments, language used to encode spatial relations, and if so, better classified as thinking “with” language e.g., meddler or augmenter (Wolff & Holmes, 2010)
