Outlines Effects of Lexical Spatial-Temporal Metaphor l Introduction - - PDF document

Effects of Lexical Spatial-Temporal Metaphor

• n Mandarin and Cantonese Speakers'

Temporal Conceptualizations

Pan Zhu, Yu-Yin Hsu The Hong Kong Polytechnic University The 21st Chinese Lexical Semantics Workshop

1

Outlines

l Introduction l Experiment 1: Survey l Experiment 2: Implicit-Processing Experiment l Discussion and Conclusion

2

Introduction

• People's language systems and cultures may influence their

temporal conceptualizations (Boroditsky, 2001; 2008; 2013; Fuhrman & Boroditsky, 2011; Liu & Zhang, 2009).

l Language systems: The spatial-temporal metaphors l Cultures: the reading-writing habits

• Some research does not support this view (Chen 2007).
• èThese studies compared a word (in a temporal phrase) with a

morpheme (in a temporal compound), e.g.,

l the good times ahead of us vs. qian-nian (front-year) ‘last year’

3

Introduction

• The differences of morphemes in temporal compounds between

Mandarin and Cantonese：

l ‘last year'

• --- qu-nian ‘backward-year’ (Mandarin)
• --- soeng-nin ‘up-year’ (Cantonese)

l ‘next year’

• --- ming-nian ‘tomorrow-year’ (Mandarin)
• --- haa-nin ‘down-year’ (Cantonese)
• Q1: Do such morpheme-spatial differences influence speakers’

conceptualization of time?

4

Introduction

• Hong Kong Cantonese speakers read vertical and horizontal texts.
• Most Beijing Mandarin speakers only read horizontal texts.
• Vertical texts in Chinese printed transversely from right to left;

horizontal texts printed transversely from left to right.

• Q2: Do different human constructs (in reading and writing

directions) influence speakers’ conceptualization of time?

從 右 至 左 直 式 書 寫 横 式 書 寫 從 左 至 右

5

Experiment 1: Survey

• Tow existing Cantonese corpora have only 4 temporal expressions.

è Create a comparable data set of Cantonese-Mandarin through surveys

• The survey was distributed to participants in hardcopy.

Traditional characters for Cantonese participants, and simplified characters for Mandarin participants.

• Participants
• 69 Beijing Mandarin speakers, aged 18-30 and living in

Beijing; none had a good command of Cantonese.

• 32 Hong Kong Cantonese speakers, aged 18-30 and living in

Hong Kong; none had learnt Mandarin before age six or had received formal Mandarin education.

6

Experiment 1: Survey

• Survey Materials

Three parts of elements:

l 4 spatial-metaphorical terms: 前 ‘front’, 后 ‘back’, 上 ‘up’, 下 ‘down’ l 11 Time units:

世纪 ‘century’, 年 ‘year’, 季度 ‘quarter’, 月 ‘month’, 星期 ‘week’, 周 ‘week’, 天 ‘day’, 日 ‘day’, 小時 ‘hour’, 分鐘 ‘minute’, and 秒 ‘second’.

l Inserted elements:

measure word: 個 ge; numeral” 一 ‘one’ and 半 ‘half’.

7

Experiment 1: Survey

• Survey Procedure

All participants were instructed to choose (√) which of the four spatial-metaphorical terms was (the most) appropriate to each of the six combinations and they were allowed to choose more than one answer.

世纪 shiji 个世纪 ge shiji 半世纪 ban shiji 一世纪 yi shiji 半个世纪 ban ge shiji 一个世纪 yi ge shiji 前 qian 后 hou 上 shang 下 xia

8

Experiment 1: Survey

Analyses

• We obtained the “vertical preference value” by subtracting the

value of the percentage of sagittal metaphors selected by the participants on each item from that of the vertical metaphors they selected. Results (1): General preference

• Mandarin speakers use sagittal

metaphors more frequently

• Cantonese speakers use vertical

metaphors more frequently

9

Experiment 1: Survey

Results (2): Differences with respect to the length of time unit

l Longer time units (ranging from 世纪 'century' to 星期 'week',

except for 年 'year'): both groups showed vertical preferences.

l 年 'year’: the Mandarin group preferred sagittal metaphors and the

Cantonese group showed no clear preference. èThis might be because Cantonese speakers commonly use the vertical metaphors for ‘year’.

l Shorter time units (ranging from 秒 ‘second’

to 小时 ‘hour’, except for 天 'day’): the Cantonese group prefer vertical metaphors while Mandarin group had no clear preference.

l 天 'day’: both groups had sagittal preferences.

10

Experiment 1: Survey

Results (3): some system-internal differences Mandarin speakers’ choices seem to be more phrasal compositional while Cantonese speakers’ were more based on the temporal concepts.

l When 半 'half’ occurred with the temporal expressions, Mandarin

group preferred sagittal metaphors for most of the terms, whereas Cantonese group’s preference became more vague but still showed vertical preference.

l When 一 'one’ occurred with the temporal expressions, Mandarin

group preferred the sagittal metaphors for ‘hour’, while this did not affect Cantonese groups’ judgements.

11

Experiment 2: Implicit-Processing Experiment

• The key-pressing-based judgment tasks devised by Fuhrman (2011) è To test

whether there are differences between the two groups’ temporal conceptualizations.

• The mini USB keyboard was installed on a tripod mount and could be rotated
• vertically. The middle key was masked with a blue sticker. The key with a black

sticker expressed “earlier” concept and the key with a grey sticker expressed “later” concept.

• The materials were 40 triplets of pictures.

12

Experiment 2: Implicit-Processing Experiment

Participants

• 30 Beijing Mandarin speakers, aged 18-30, university students in Beijing; none of

whom had participated in our survey. None had a good command of Cantonese, and all rarely read vertical texts.

• 30 Hong Kong Cantonese speakers, aged 18-30 university students in Hong Kong;

none of them had participated in our survey. None of them had learnt Mandarin before age six, but were free of language-understanding and expression barriers.

13

Experiment 2: Implicit-Processing Experiment

Experimental Procedure

• In each trial, a picture presenting a middle time point (e.g., an apple being eaten)

was presented on the screen, followed by a picture showing either an earlier time point (e.g., a whole apple) or a later time point (e.g., an apple core). Participants needed to press the key corresponding to judge whether the second picture showed an earlier or a later conceptual time point than the first picture.

• All participants completed four test blocks, each consisting of 40 trials, two blocks

covering the transverse axis (left-earlier/right-later key mapping and its opposite) and two blocks on the vertical axis (top-earlier/bottom-later key mapping and its

• pposite). The order of the four blocks was counterbalanced with each item

appearing once in each axis.

14

Experiment 2: Implicit-Processing Experiment

Data Analysis

• Linear mixed effects models were created using log-transformed reaction

times (log RT) with the lme4 package (Bates et al., 2015) in R (R Team, 2016).

• The significance of the main effect was evaluated by likelihood-ratio

testing, and post-hoc Tukey’s comparisons were conducted using the multcomp package (Hothorn et al., 2008) in R.

15

Experiment 2: Implicit-Processing Experiment

Results

• Both groups processed transverse-axis items with canonical key-

mapping significantly faster than items on the same axis with non- canonical key-mapping (ps<.001).

• Both groups processed transverse-axis items in the canonical condition

significantly faster than vertical-axis items in the canonical condition (ps<.001).

• Cantonese speakers were faster to process vertical-axis items in the

non-canonical condition than Mandarin speakers (p = .042).

16

Di Discussion and Conclusion

Q1: Do morpheme-spatial differences influence speakers’ conceptualization of time?

l Yes. Cantonese uses vertical metaphor in temporal compounds more

than Mandarin does, and in our experiment, the Cantonese group showed shorter reaction time in trials related to the vertical axis. Q2: Do different human constructs (in reading and writing directions) influence speakers’ conceptualization of time?

l Cantonese speakers often read vertical texts, and they were faster to

make judgements during the task on trials of the vertical axis. è We are inclined to conclude that Cantonese speakers employed vertical metaphors in temporal expressions more frequently than Mandarin speakers do, and that such lexical differences between the two groups influenced these speakers’ mental conceptualizations of time.

17

Ac Acknowl wledgments: This study was supported by the Faculty of Humanity in Hong Kong Polytechnic University and the HK PolyU-PKU Research Centre on Chinese Linguistics (RP2U2). We thank Stephen Politzer-Ahles and Chu-Ren Huang for their comments and suggestions on the earlier stages of this study. We also thank Xia Wang for the technical support.

Thank you!

18