Temporal frames of reference in three Germanic languages: Individual consistency, interindividual consensus, and cross-linguistic variability

Abstract

A task like “moving a meeting forward” reveals the ambiguity inherent in temporal references. That speakers of U.S. English do not agree on how to solve it is well established: Roughly one half moves the meeting futurewards, the other half pastwards. But the extent to which individual speakers, rather than groups of speakers, consider such phrases as ambiguous has not been scrutinized. Does the split in readings result from a lack of intraindividual consistency or from a lack of interindividual consensus? And how specific is U.S. English in this regard when compared to other closely related Germanic languages? Based on a taxonomy of spatiotemporal frames of reference (FoRs), we conducted two experiments with speakers of Swedish, U.S. English, and German to assess individual preferences for temporal FoRs, intra- and cross-linguistic variability, consistency and long-term stability of these preferences, and possible effects of priming a spatial FoR. The data reveal cross-linguistic differences, both in terms of which temporal FoRs speakers prefer (the absolute FoR in Sweden, the intrinsic FoR in German, and both of these in the US) and in terms of the extent to which these preferences are shared and stable (high consensus and consistency in Sweden and Germany, and low consensus and partial consistency in the US). Overall, no effect of spatial priming was observed; only speakers of U.S. English with a baseline preference for the absolute temporal FoR seemed to be susceptible to spatial priming. Thus, the assumption that temporal references are affected by spatial references is only weakly supported.

Keywords:

• Frames of reference
• Time
• Space
• Cross-linguistic comparison
• Priming

For collecting the data of Experiment 1, we thank Laura Henn (Swedish data), Giovanni Bennardo (US data), and Gregory Kuhnmünch (German data); for collecting the US data of Experiment 2, we thank Lisa Hüther and the lab of Judith F. Kroll (Pennsylvania State University). We are also grateful to Sarah Mannion de Hernandez for proof-reading.

This research was supported in part through a Heisenberg Fellowship from the Deutsche Forschungsgemeinschaft DFG to Andrea Bender [grant number Be 2451/8-1,2] and through a grant for the project “Spatial referencing across languages: Cultural preferences and cognitive implications” to Andrea Bender [grant number Be 2451/13-1] and Sieghard Beller [grant number Be 2178/7-1].

Notes

¹As our claim that events can be assigned a front remains controversial, we recently conducted a pilot survey with 145 native speakers of German, in which we asked where the front of an event is (with four response options being given: beginning, end, something else, or does not exist). While 33% responded that events do not have a front, 65% did assign front to events: 58% to their beginning, and 7% to their end.

²In total, 388 students had participated in this study: 87 in Sweden, 157 in the US, and 144 in Germany. Of these, 22 were excluded (5 from the Swedish sample, 15 from the US sample, and 2 from the German sample), as they either indicated to be bi- or multilingual or did not indicate their mother tongue. In Sweden and Germany, most students have some proficiency in English, as English typically is taught as a foreign language at school. On the other hand, in the US, only few students can be assumed to have knowledge of German and/or Swedish.

³Our findings also speak against a futurewards preference for those tasks that involve clock time, and this seems to be in conflict, at first glance, with the findings reported in Lai and Boroditsky (2013). However, the two tasks differ in at least one crucial aspect: Whereas our participants were asked to move forward an event that was scheduled for a certain time, the participants in Lai and Boroditsky (2013) were asked to move forward the clock itself. Clocks—and especially those that have hands—can be regarded as having a spatial moving direction, which might prompt people to choose this direction, whereas an event such as a meeting may not be regarded as having such an inherent spatial moving direction.

⁴In total, 468 students had participated in this study: 151 in Sweden, 140 in the US, and 177 in Germany. Of these, 27 were excluded (9 from the Swedish sample and 18 from the German sample), as they either indicated to be bi- or multilingual or did not indicate their mother tongue.

⁵Numbers of participants in the experimental conditions were as follows: In Sweden 34 absolute, 34 intrinsic, 37 relative, and 36 control; in the US 32 absolute, 27 intrinsic, 31 relative, and 41 control, and in Germany 40 absolute, 39 intrinsic, 38 relative, and 38 control.