Content uploaded by Feifei Han
Author content
All content in this area was uploaded by Feifei Han on Sep 09, 2021
Content may be subject to copyright.
1
Language expansion in Chinese parent-child mealtime conversations: Across
different conversational types and initiators
Ling Shenga
Wenming Donga
Feifei Hanb
Shiming Tonga
Jiangbo Hua*
aHangzhou College of Preschool Education, Zhejiang Normal University, China
ORCID: 0000-0001-5471-7689
bOffice of Pro-Vice-Chancellor (Arts, Education and Law), Griffith Institute for
Educational Research, Griffith University, Brisbane, Australia.
ORCID: 0000-0001-8464-0854
* Corresponding author: Jiangbo Hu
1108 Gengwen Road, Xiaoshan District, Hangzhou
Zhejiang Normal University, China
jiangbo.hu@zjnu.cn
Acknowledgements: This study was funded by Zhejiang Normal University’s CCELE
research grant. We acknowledge the contributions of the participating families who
generously shared their information with us and the organizations who assisted the data
collection in collaboration with us.
Disclosure statement: No financial interest, benefit or conflict of interest has arisen
from any author’s participation in this research project or the resultant manuscript.
2
Language expansion in Chinese parent-child mealtime conversations: Across
different conversational types and initiators
This study examined the distribution of language expansion in parent-child (preschool
aged) mealtime conversations in 30 Chinese middle-class families. The conversations
were categorized into four types: contextualized & conflicted, contextualized & non-
conflicted, decontextualized & conflicted, and decontextualized & non-conflicted. The
language expansions were analysed using the systemic functional linguistic theory
related to cohesive patterns in language expansion: elaborations, extensions, and
enhancements. While the parents dominated the conversations generally, the children
were active contributors, initiating over one quarter of the conversations. Initiation had
an impact on the distribution of the conversational types: the proportions of
contextualized & non-conflicted conversations was significantly higher in child-initiated
conversations. The contextualized & conflicted conversations accounted for a higher
proportion in parent-initiated conversations. It was the conversational type rather than
initiation, which had an effect on the distribution of language expansion patterns. The
least occurring decontextualized & conflicted conversations generated the most
extensions. The frequently appeared contextualized & non-conflicted conversations,
however, produced fewest expanded messages. The implications from the findings for
promoting high-quality mealtime conversations conducive to children’s language
learning are discussed.
Keywords: mealtime conversations; conversational types; language expansion;
Chinese families; Chinese children
3
Introduction
Research suggests that family mealtime conversations can be regarded as a critical
context for children’s language learning and social progress (Bova and Arcidiacono
2013; Ferdous et al 2016; Fruh et al 2011; Hu et al 2019). Mealtime talk includes both
adult-adult and adult-child conversations in which the richness of topics and the
complexity of language use may be different from the instance of adult-child
interactions in other situations. Parents as skilful language users can create extensive
opportunities for children’s language learning with sophisticated vocabulary and
complex syntactic structures in their dinner table talk (Snow and Beals 2006; Busch
2017). The well-timed turn-taking, the maintenance of one’s turn and the cohesive links
with previous talk presented in the family members’ talk facilitate children’s language
practices and improve children’s language expression as well as social-cognitive
abilities (Arcidiacono and Bova 2015; Busch 2012).
Socioeconomic status (SES) and cultural background of families are widely
acknowledged as an important factor affecting the style of dinner table conversations.
In mealtime talk, middle-class parents are more likely to use expanded genres (e.g.
narrative, explanatory, or argumentative talk) to discuss decontextualized topics at the
dinner table compared to those from a disadvantaged background. These expanded
discourses are often characterized by having low-frequency words, decontextualized
topics, and justification discourses, which are conducive to children’s language
development (Arcidiacono and Bova 2015; Demir et al 2015; Rowe 2013). Middle-class
parents in these studies refer to parents who obtained university degrees or have a
professional job. Apart from SES, cultural background has also been shown to influence
mealtime talk regarding communication roles and conversational topics and styles
(Aronsson and Gottzen 2011; Galatolo and Caronia 2017). In Western culture, children
4
are typically encouraged to join the conversations with adults, whereas in some other
cultures, children are expected to keep quiet while adults talk (Rydstrom 2003; Fiese et
al 2006). Even within Western cultures, the features of mealtime talk differ from culture
to culture. For instance, Aukrust (2002) reported that the genres of mealtime
conversations in Norwegian families were different from that in in American families.
Norwegian families produced more narrative talk on social practices and events at
school whilst American families generated more explanatory talk for reasoning social
affairs. These studies indicate that mealtime conversations vary individually by
socioeconomic status and by culture.
To date, most mealtime talk research is undertaken in Western cultures and little
has been done in Chinese families, especially the families in Mainland China. It is
widely acknowledged that the manners of Chinese adult-child interactions differ
substantially from those of Western cultures (Dai et al 2020; Doan and Wang 2010). In
addition, much of the existing research predominantly focuses on adult language use in
mealtime talk, with children’s language use at the table dinner being largely ignored.
However, children are not merely passively affected by the language environment but
are active contributors to the social interactions they are involved in ( Halliday 2004).
To address these less researched areas, this study examines language use of both parents
and children in Mainland Chinese families who speak Mandarin (Chinese families
hereafter) mealtime conversations. Specifically, we focus on language expansions in
expanded genres (e.g. narrative or explanatory talk) that regards the feature of ideas
being exchanged spontaneously and cohesively in conversations, as they have extensive
pedagogical functions for children’s language learning (Bohanek et al 2009; Busch
2017; Snow and Beals 2006). Results of this study would increase the cultural diversity
5
in the mealtime research field and enhance our understanding about the language
environment of Chinese families that is established by both parents and children.
Learning opportunities in mealtime conversations for children
Adopting a sociolinguistic perspective, research that investigates the pedagogical
functions of mealtime conversations emphasizes the influence of language interactions
on children’s learning experiences ( Halliday 2004). In particular, the expanded genres
in mealtime conversations (e.g. narratives, explanations, or arguments) containing
decontextualized information, are said to be favourable, as they create substantial
language learning occasions for children to develop vocabulary knowledge, expressive
discourses, and early literacy skills. (Aukrust 2002; Fruh et al 2011; Rowe 2013). Snow
and Beals (2006) examined the proportion of parents’ narrative and explanatory talk in
their interactions with their preschool children (four or five years) at the dinner table
and identified positive correlations between the quantity of the talk and their children’s
performance in The Peabody Picture Vocabulary Test in grade six. Bohanek et al
(2009), however, explored the influence of parents’ narrative talk at the dinner table on
children’s social-cognitive learning experiences. The researchers found that mothers
and fathers play different roles in supporting their children to gain understanding in life
events of remote and recent. The fathers’ narrative talk focused on day-to-day problem
solving whereas the mothers’ talk paid attention to shared family history or rationale of
social affairs, all of which assisted the children to create coherence over their
experiences relating to their emotional bond, self-awareness, social understanding and
logical thinking.
A handful of studies also revealed the potential benefit of arguments in mealtime
conversations for children’s language and social-cognitive development (Arcidiacono
and Bova 2015; Bova and Arcidiacono 2014; Busch 2012). This type of expanded talk
6
drives children to react quickly to address their parents’ challenging proposals for
defending their own viewpoints, during which their expressive language and the social-
cognitive skills for providing appropriate negotiations are practiced and strengthened. In
dinner table arguments, conflicting parent-child conversations are frequently observed
(Bova 2019; Bova and Arcidiacono 2014). These conflicting conversations initiated by
parents tend to relate to contextualized issues, such as children’s table manners or food
intake; whereas the conflicts initiated by children are more likely to involve
decontextualized questions, like information seeking outside family issues, which has
been highlighted to be important in children’s language and social development (Bova
2019). These studies suggest that children are active participants in mealtime
conversations and are able to enrich the content of mealtime conversations significantly.
Much of the existing research examines the pedagogical functions of parent-
child mealtime conversations separately in two dimensions: 1) whether the mealtime
conversation has contextualized information; and 2) whether the conversation involves
a conflict or not. The present study sets to investigate the parent-child conversations
considering the interaction between the two dimensions, which forms four types of
conversations: namely contextualized & conflicted, contextualized & non-conflicted,
decontextualized & conflicted, and decontextualized & non-conflicted. This kind of
categorization extends the existing research through the examination of the nuances of
parent-child mealtime conversations in the specific conversational types that were rarely
reported.
Parent-child communication in Chinese families
Although there is little research on parent-child mealtime talk in Chinese families, a
number of studies have investigated communication styles in Chinese families (from
Mainland China and overseas Chinese communities). Inconclusive results have been
7
reported with regard to the levels of parental authoritativeness and child-centeredness in
Chinese parents’ interactions with children (Guo 2016; Luo et al 2013). Such mixed
results may well reflect the sociocultural change in China shifting from traditional
Chinese morals (e.g. Confucianism) to a mixture with Western values due to
globalization (Guo 2016; Luo et al 2013). Earlier studies of the communication styles in
older generation Chinese families tend to report the predominate “Jiazhang” role (the
leading role) of Chinese parents in a family. For instance, in an earlier study, Wang
(2001) compared the discussion of Chinese mothers and American mothers with their
children about past emotional events. Chinese parents were found to use much more
directive language with less emotional expressions compared to their Western
counterparts. In more recent studies, however, the younger generation of Chinese
parents were reported to show active and open styles in their interactions with their
young children for promoting children’s autonomy and agency (Li, 2020; Hu and Torr
2016).
Apart from parent-child communication styles, research has also been conducted
to examine the impact of Chinese parent-child communication on children’s cognitive
and socio-emotional development based on the notion that language plays a vital role in
fulfilling individual’s social and cognitive developmental progress (Halliday, 2004).
Zhang (2018) investigated 51 Chinese mothers’ interactions with their three-year-old
children and found that the mothers’ interactional behaviors (e.g. language use in praise
or criticism) were not only associated with children’s language performance but also
their emotional understanding. From the perspective of language functions, Hu and Torr
(2016) explored the use of reasoning talk in Chinese mother-child daily conversations.
The researchers reported that when explaining affairs or directing children’s behaviors,
Chinese mothers’ reasoning talk with logical and cooperative principles may foster
8
children’s understanding of the natural laws of physical world and the social rules in
Chinese society. These studies demonstrate that through spontaneous daily
conversations, parents may significantly shape their children’s way of thinking and
manner of social interactions.
Like the mealtime talk research, the existing studies of Chinese parent-child
communication extensively discussed parents’ communication styles or parents’
language use, yet little attention was paid to children’s language contribution to the
communication. In some socioeconomically disadvantaged Chinese families where
parents have few conversations with children, the children play active roles in initiating
conversations (Ji and Zhang 2020). The relations between children and adult-child
communication are bi-directional. Hence, studies neglecting children’s role in the
interactions could only reveal an incomplete picture of the complexity of parent-child
interactions. To address this issue, the present study adopted a dual perspective by
including both the Chinese parents and children’s language use.
A theoretical framework for analysing language expansion in parent-child
conversations
This study focuses on expanded genres in parent-child conversations that contain
extensive language expansions. The analysis of the language expansion reveals the flow
of informational exchange between parents and children that reflects the speakers’
metalinguistic sense in maintaining the cohesion of the conversation for “meaning
making” in collaboration (Eggins 2004). Halliday (2004) argues that language learning
includes three levels: learning language that refers to language knowledge (e.g. lexicon
and syntactic knowledge); learning through language that concerns the interpersonal
social roles in an interaction (e.g. modality); and learning about learning that relates the
use of language appropriating for the specific instance in interactional contexts (e.g.
9
theme choice or cohesion maintenance). The analysis of language expansion enables
identification of language learning opportunities embedded in mealtime conversations
from perspectives of “learning about language”.
The present study draws on the theory of “mode” in Systemic Functional
Linguistics (SFL). Mode concerns the cohesion of language that is suited in an instance
of social interaction (Halliday and Matthiessen 2004). According to Halliday and
Matthiessen, there are three main conjunctive relations (expansion patterns, which is
used hereafter) for maintaining the coherence in language use: elaboration, extension,
and enhancement. Table 1 displays the descriptor, representative lexical expressions,
and the example sentences of the three expansion patterns. To some extent, the
expansion patterns are associated with conversational genres (Eggins 2004). For
example, extensions are widely used in a narrative talk whereas enhancements are more
likely to appear in argumentative discourse.
Table 1. The three expansion patterns
Expansion
patterns
Descriptors
Representative
lexical expressions
Examples
Elaboration
A message restates
or clarifies the
previous message in
the text
in other words, I
mean, for instance,
to illustrate, in fact,
actually…
Today is a special
day, actually, it is the
day for eating
Zhongzi (a special
food for a Chinese
traditional festival).
Extension
A message adds new
information to or
changes the meaning
and, also, moreover,
in addition, nor, but,
yet, instead…
You said Dingding
snatched your toy,
10
of the previous
message
and you told your
teacher, right?
Enhancement
A message extends
the previous
message in terms of
dimensions
including time,
comparison, cause,
condition, or
concession
then, next,
afterwards, just
then, soon, until,
because, therefore,
otherwise,
despite…
I like doing math
because I need to
calculate many things
in the future.
The present study
Because of the pedagogical functions of mealtime conversations, in particular, the
expanded genres of mealtime talk, the present study aimed to examine: 1) the
distribution patterns of different types of Chinese parent-child mealtime conversations;
and 2) the distribution patterns of language expansions in Chinese parent-child
mealtime conversations. This study is significant in three aspects. Firstly, the present
study will include both parents’ and children’s language in the mealtime conversations.
This study will make distinction as to whether the conversation is initiated by parents or
children, and to see the extent to which differences in the initiation influence the
distribution patterns of conversations and language expansions in the conversations.
Secondly, as different types of conversations may be affected by the initiator, and
language expansion patterns may also appear differently in different types of
conversations and by initiator. Hence, the present study will examine the effect of
11
conversational types on the language expansion patterns that is rarely explored in the
existing research. Thirdly, as culture has been shown to be an influential factor
impacting parent-child conversational styles, the present study will examine parent-
child mealtime conversations in Chinese families, as the majority of research has been
conducted in Western countries, which differ from China in terms of culture.
Specifically, the following research questions are addressed:
1. What are the distribution patterns of: a) the four types of conversations; and b) the
four types of conversations by initiation; in Chinese parent-child mealtime
conversations?
2. What are the distribution patterns of language expansions: a) in the total
conversations; and b) by initiation, conversational types, and the interaction between
initiation x conversational types; in Chinese parent-child mealtime conversations?
Materials and Methods
Participants and data generation
The study is part of a larger project investigating Chinese children’s early language
experiences (CCELE) at home and in childcare centres. The participants of the study
were 30 Chinese middle-class families recruited from a local preschool in Deqing,
Zhejiang Province, China. The focus on middle-class families is based on the
consideration that the main objective of this study is to reveal pedagogical functions in
expanded genres in mealtime talk. Previous research suggests that this type of language
is more likely to appear in middle-class families (Bohanek et al 2009; Bova and
Aridiacono 2013). The recruitment of the participants was targeted on those parents
who 1) had professional jobs; 2) at least one of the parents had received university
education; 3) had only one preschool-aged child; and 4) speaking Mandarin at home.
This group of families also represent the fast-growing group of population in Chinese
12
small cities/towns. Ethical approval for the larger project was obtained from Zhejiang
Normal University and the signed written consent forms were collected from the
participating preschools and families. The parents were informed about the nature of the
study and were told to record their dinner table interactions with their children using a
mobile phone by themselves at any time. The parents were encouraged to discuss with
their children about this video recording meal in advance and anecdotal data showed
that the children were aware of this event. The recorded video clips were sent to the
preschool and then collected by the research assistant.
The original clips were between 20 to 35 minutes. As the study targeted the
mealtime conversations of all the family members (mother, father, and the child), the
parts of table setting or without the participation of all the family members, or the
parents not sitting with their children were excluded. The exclusion resulted in
approximately 15-minute video for each family.
Coding of the data
The selected video clips were transcribed verbatim and were coded using a five-step
scheme as shown below. The development of the coding scheme was grounded on the
relevant theories of SFL that have been widely adopted in language studies in early
childhood education, including studies with Mandarin-speaking Chinese parents, like
the current participants (Hu et al 2019; Hu and Torr 2016).
• 1) Identification of messages: The sentences were broken into messages. One
message is defined as the smallest semantic unit, which includes an implicate or
explicate subject and a verb (Hasan 1996).
• 2) Identification of conversations: In this study, one conversation was defined as
the parents’ and the child’ turn takings which focused on one topic and lasted at
least four turns.
13
• 3) Classification of conversations into four types, namely contextualized &
conflicted, contextualized & non-conflicted, decontextualized & conflicted, and
decontextualized & non-conflicted conversations: The information of a
conversation was examined on: 1) whether it was contextualized with here-and-
now food eating issues or decontextualized information away from the dinner
table; and 2) whether it had conflicts between the parents and the children on the
topic or not. The examples of the four types of conversations are listed in Table 2.
Table 2. Four types of conversations
Types of conversations
Examples
Contextualized & conflicted
Child: I cannot eat any more
Mother: Just little bit rice and dishes
Child: I really cannot eat any more
Mother: It is not good if you don’t eat the
food.
Contextualized & non-conflicted
Mother: Hi, can I give a fish? Mind the
small bones.
Child: Ok, I can see the small bones.
Mother: OK
Child: I find a big bone
Decontextualized & conflicted
Mother: I think dad can take you out to play
when you come back (from the preschool).
Child: We go to the Yager Zoo?
Mother: It would be very hot if we go the
the Yager Zoo. How about “Sea World”?
14
Child: No, I don’t like there.
Decontextualized & non-conflicted
Father: Xiangru (the child’s name) Do you
want to eat fried dumpling for breakfast?
Mother: Did you have fried dumpling in the
morning?
Child: Yishi Tample, we had it around Yishi
Tample
Mother: Wow, that is a famous restaurant.
Child: It is delicious.
•
• 4) Identification of language expansions: The messages in each conversation were
examined one by one to identify whether they belonged to language expansions,
which were defined as a message expanding the meaning of the previous messages.
• 5) Categorization of expansion patterns: Each identified expanded message was
categorized as one of the three expansion types, namely elaboration, extension or
enhancement.
Inter-coder reliability
The data coding was processed by one researcher independently in Excel. To ensure the
reliability of the coding, a second coder randomly selected and coded the data of six
families out of 30 families, representing 20.0% of the total data. The inter-coder
reliability – Cohen’s Kappa coefficients were calculated. The coefficients were .83 for
initiation, .80 for conversational types, and .76 for expansions. According to McHugh
(2012), all the values of the Cohen’s Kappa coefficients showed substantial agreement.
Data analysis
15
Individual conversation rather than individual family was used as a case in the data
analyses, because the main aim of the research was concerned with revealing the
generic distribution patterns of language expansions in conversations across families
rather than detecting variations of such patterns between families by their different
characteristics. To minimize the impacts of individual differences of families, their SES
background, location of the families, parents’ education and jobs, language speaking at
home, formation of the families were all controlled for in the participants recruitment
(see Participants for details).
To answer research question 1a), a series of one-sample proportion tests were conducted
for pairwise comparisons of the proportions of the four types of the conversations. For
research question 1b), a 4 (conversational type) x 2 (initiation) cross-tabulation was
conducted. Similarly, to provide an answer to research question 2a), one-sample
proportion tests were performed for pairwise comparisons of the proportions of the
three types of language expansions in total conversations. With regard to research
question 2b), a 2 (initiation) x 4 (conversational type) MANOVA was used, with the
three types of language expansions as dependent variables. The data analyses were
carried out in IBM SPSS 25.
Results
Results of distribution patterns of the total and the four types of conversations by
initiation
Table 3 presents the descriptive statistics of the total and the four types of conversations
by initiation. Altogether 6,007 messages generated 633 conversations, of which parent-
initiated conversations comprised 463 (73.1%) and child-initiated conversations
comprised 170 (26.9%). A one sample proportion test showed that parents initiated a
16
significantly higher proportion than children (z=26.21, p<.01). The results of the one-
sample proportion tests for the pairwise comparison of the four types of conversations
are displayed in Table 4. Viewed together, these results demonstrate that irrespective of
the initiator, decontextualized & non-conflicted conversations (42.0%) accounted for the
highest proportion, followed by contextualized & non-conflicted conversations (36.0%),
which in turn had higher proportion than contextualized & conflicted conversations
(14.75%). The decontextualized & conflicted conversations (7.4%) had the lowest
proportion amongst the four types of conversations.
Table 3. Descriptive statistics of the total and the four types of conversations by
initiation
Variables
parent-initiation
child-initiation
M
SD
M
SD
total conversations
15.43
4.16
5.67
3.01
contextualized & conflicted conversations
2.73
1.86
0.33
0.55
contextualized & non-conflicted conversations
4.93
2.36
2.67
1.77
decontextualized & conflicted conversations
1.27
1.42
0.30
0.65
decontextualized & non-conflicted
conversations
6.50
4.29
2.37
2.08
Table 4. Pairwise comparison of the proportions of the four types of conversations
conversational types (number, proportion)
z
p
1 contextualized & conflicted conversations (92, 14.5%)
1<2
15.36
.00
1>3
-5.07
.00
1<4
19.65
.00
2 contextualized & non-conflicted conversations (228, 36.0%)
2>3
-14/99
.00
17
2<4
3.15
.00
3 decontextualized & conflicted conversations (47, 7.4%)
3<4
33.26
.00
4 decontextualized & non-conflicted conversations (266, 42.0%)
---
---
---
As to the distribution pattern of the four types of the conversations, the results of
the 4 (conversational types) x 2 (initiation) cross-tabulation are presented in Table 5.
The conversational types was significantly associated with the initiation (χ²(3)=21.26,
V=.18, p<.01). Specifically, the contextualized & conflicted conversations accounted for
a significantly higher proportion in parent-initiated conversations (17.7%) than in child-
initiated conversations (5.9%). The contextualized & non-conflicted conversations took
up a significantly higher proportion in child-imitated conversations (47.1%) than in
parent-initiated conversations (32.0%). For the other two types of conversations, no
significant differences were found between the proportions that accounted for in parent-
and child-initiated conversations.
Table 5. Association between conversational types and initiation
conversational types
initiation
parent
child
total
contextualized & conflicted
conversations
Count
82a
10b
92
% within initiation
17.7%
5.9%
14.5%
contextualized & non-conflicted
conversations
Count
148a
80b
228
% within initiation
32.0%
47.1%
36.0%
decontextualized & conflicted
conversations
Count
38a
9a
47
% within initiation
8.2%
5.3%
7.4%
Count
195a
71a
266
18
decontextualized & non-conflicted
conversations
% within initiation
42.1%
41.8%
42.0%
total conversations
Count
463
170
633
% within initiation
100.0%
100.0%
100.0%
Note: Each subscript letter denotes a subset of categories whose column proportions do
not differ significantly from each other at the .05 level.
Results of distribution patterns of the language expansions in total conversations and
by initiation and conversational type
Table 6 shows the descriptive statistics of the three types of expansions by
conversational type and by initiation. Altogether, there were 3,443 expanded messages
in the total conversations. As shown in Table 7, for the distribution pattern of the
language expansions in the total conversations, the results of one-sample proportion
tests find that extensions took the highest proportion (67.1%), followed by the
proportion of enhancements (19.1%), which were again higher than the proportion of
elaborations (13.8%).
Table 6. Descriptive statistics of the three types of expansions by conversational type
and by initiation
variables
parent-initiated
child-initiated
total
M
SD
M
SD
M
SD
elaborations in contextualized &
conflicted conversations
0.95
0.99
0.60
0.84
0.91
0.98
elaborations in contextualized &
non-conflicted conversations
0.61
0.89
0.67
0.90
0.64
0.89
elaborations in decontextualized &
conflicted conversations
0.87
0.78
0.67
0.50
0.83
0.73
19
elaborations in decontextualized &
non-conflicted conversations
0.81
1.01
0.69
0.86
0.77
0.97
extensions in contextualized &
conflicted conversations
3.59
2.45
4.40
1.90
3.67
2.40
extensions in contextualized & non-
conflicted conversations
2.85
2.08
3.05
2.52
2.92
2.24
extensions in decontextualized &
conflicted conversations
5.11
4.56
6.22
2.22
5.32
4.21
extensions in decontextualized &
non-conflicted conversations
4.10
3.15
3.62
2.53
3.97
3.00
enhancements in contextualized &
conflicted conversations
1.26
1.54
0.60
1.27
1.18
1.52
enhancements in contextualized &
non-conflicted conversations
0.78
1.19
0.93
1.23
0.83
1.20
enhancements in decontextualized
& conflicted conversations
1.32
1.60
1.89
1.17
1.43
1.53
enhancements in decontextualized
& non-conflicted conversations
1.05
1.53
1.24
1.52
1.10
1.53
Table 7. Pairwise comparison of the proportions of the three types of language
expansions in total conversations
language expansion types (number, proportion)
z
p
1 elaborations (474, 13.8%)
1<2
90.68
.00
1<3
9.02
.00
2 extensions (2310, 67.1%)
2>3
-59.95
.00
20
3 enhancements (659, 19.1%)
---
---
---
The results of the 2 x 4 MANOVA show that the main effect of initiation was
not significant in all the three types of language expansions. The main effect of
conversational type, however, had significant effects on two out of three language
expansions, namely extensions (F(3, 625)=9.96, p<.01, partial η2=.05) and
enhancements (F(3, 625)=3.17, p<.05, partial η2=.02). There was no significant
interaction effect between initiation x conversational type on all the three language
expansions.
For the conversational type on extensions, the post-hoc pairwise comparisons
reveal that the decontextualized & conflicted conversations generated the most
extensions (M=5.32, SD=4.21), followed equally by contextualized & conflicted
(M=3.67, SD=2.40) and decontextualized & non-conflicted (M=3.97, SD=3.00)
conversations. The contextualized & non-conflicted produced the fewest extensions
(M=2.92, SD=2.24). For enhancements, the pairwise comparisons show that
decontextualized & conflicted (M=1.43, SD=1.53) and decontextualized & non-
conflicted conversations (M=1.10, SD=1.53) generated significantly higher
enhancements than contextualized & non-conflicted conversations (M=0.83, SD=1.20).
However, enhancements did not differ between contextualized & conflicted
conversations and the other three types. Nor did it differ between decontextualized &
conflicted conversations. These results were visualized in Figure 1.
21
Figure 1. The distribution of language expansions by conversational types
Discussion
This study explores the language expansions in parent-child conversations at the dinner
table drawing on the stances of both parents and children’ language use in a group of
middle-class Chinese families. Previous research identified pedagogical functions of
expanded genres at the dinner table, yet the details about how such genres would offer
language learning opportunities for children are less clear. Grounded in SFL theories,
this study extends existing research in the field as it reveals the cohesive patterns in the
expanded messages, which offer great potential for children to develop language skills
of how to maintain language cohesion by constantly constructing meanings in
conversations. The findings of this study also add to the literature of how Chinese
children’s language learning experiences are shaped in the family daily routine (at the
dinner table) in Mainland China, which is an under-researched area.
The distribution of four types of conversations in Chinese parent-child mealtime
conversations
22
The findings relating to the conversational types suggest that the parents dominated in
all the four types of the conversations and generated a significantly higher proportion of
conversations than their children did. This finding corroborates the studies showing
parents’ leading position at the dinner table that create the conversations providing a
supported and challenging situation for their children (Bova and Arcidiacono 2014;
Brumark 2006). However, children’s contribution to the mealtime conversations was
also noticeable in that they generated more than one quarter of all the conversations,
which demonstrates the young children’s active role in creating the language
environment at home.
The participating families’ dinner table talk is presented with a range of
conversations containing extensive language expansions, indicating the dynamic
informational exchange among the family members on shared topics. Among the four
types of conversations, decontextualized & non-conflicted conversations occurred most
frequently and accounted for the highest proportion in the parent-initiated
conversations. This result signifies the pedagogical functions and harmonious parent-
child relationships embedded in these families’ conversations. This finding may be
attributable to the background of the participating parents, who are well-educated and
professionals (middle-class). It is widely acknowledged that the well-educated middle-
classed parents tend to use narrative or explanatory discourses which focus on
decontextualized information (Bohanek et al 2009; Busch 2017). Our results are also
consistent with previous research on Chinese parent-child communication that well-
educated younger generation of Chinese parents are reported to have warm and trusting
relationships with their children and have open communication by significantly
involving children in family communication.
23
The prevalence of contextualized & non-conflicted conversations, which ranks
the second most among the four conversational types, takes a substantial proportion of
more than one third of total conversations. The proportion of this conversational type in
child-initiated conversations amounted to almost half, which was significantly higher
than that in parent-initiated conversations. In such conversations, the children showed
positive attitudes or practices relating to food eating issues that were supported by the
parents. Ranked as the third among the four conversational types, as the contextualized
& conflicted conversations, however, showed completely opposite features (children’s
intense resistance) from those in contextualized & non-conflicted conversations. The
proportion of contextualized & conflicted in the parent-initiated conversations (17.7%)
was nearly three times than that in the child-initiated conversations. This result concurs
with Bova’s (2019) study that revealed parents’ initiation of conflicting conversations
often concerning contextualized issues (children’s eating). Whilst the disputes between
parents and children over food eating seem to be a common issue in many cultures
(Arcidiacono and Bova 2015), Chinese parents are reported to demonstrate a high level
of anxiety relating to their children health and eating habits (Guo 2016; Hu et al 2019).
This anxiety may be rooted in the Chinese cultural faith that parents should take full
responsible for providing superb caring and learning experiences to their children so
that children can grow physically, socially and academically well (Luo et al 2013). The
enforcement for children to take more food may reflect the Chinese parents’ ideology
of “being responsible” for their children’s wellbeing, however, this practice departs
from parental authority rather than children’s real physical needs and tends to trigger
children’s resistance. Hence, this practice seems not to be in line with educational
principles for fostering children’s wellbeing.
24
The least demonstrated conversational type is decontextualized & conflicted, in
which the parents and the children argued for their standpoints for “there and then”
topics, such as Can a princess join in dragon boat competition? Considering the
children of this study are only preschool aged, it is reasonable to see the low frequency
of this type of conversations. Preschool-aged children’s argumentative skills are limited,
particularly in activity-unbound arguments, which the decontextualized information
with reasoning discourse is often required (Arcidiacono and Bova 2015). It was very
challenging for the preschool children to actively engage in such conversations.
The distribution of language expansion in Chinese parent-child mealtime
conversations
In terms of language expansions in the conversations, extensions are used mostly in the
total conversations, with elaborations and enhancements appearing much less. This
finding indicates that the parents and the children generally expanded the topics with
additional ideas on the previous messages. The findings of fewer elaborations and
enhancements might be due to the small proportions of conflicted types (regardless of
contextualized or decontextualized) in the total mealtime conversations, as the
conflicted conversations were likely to trigger elaborations and enhancements through
speakers’ frequent use of justification discourses to reconcile the conflicts. This result is
different from the findings in Bova and Arcidiacono’s studies (2013, 2014), which
reported frequent arguments occurred in parents and children’s mealtime conversations
in some Western families. The divergent results are likely to be related to differences in
Western and Chinese culture. As Chinese culture values harmonious familial
relationships and parental authorities (Luo et al 2013), some arguments could be
avoided between parents and the child.
25
Perhaps the most interesting point of this study relates to the effect of the
conversational types on the expanded messages. The type of decontextualized &
conflicted conversations occurred least, yet it contains the most extensions and also
comparatively more enhancements. This finding is supported by the research on
mealtime argument that identified the learning opportunities for children to co-construct
the “meaning making” with their parents over the conflicting topics through justification
discourses (Bova 2019). Unlike the contextualized & conflicted conversations, which
primarily focus on topics of food eating and often result in children’s unpleasant
experiences under parental authority, the arguments on “then and there” topics had more
logical reasoning language when the parents and the children tried to justify their ideas.
For example, a child reasoned that a princess could eat Zongzi (a traditional Chinese
food), because princesses are rich and can afford yummy food like Zongzi. In contrast,
although the contextualized & non-conflicted conversations appeared as the second
most frequently in all the conversations, they generated fewest expanded messages,
indicating quite limited conversational turns between the parents and the children. This
finding is coincided with language research that suggests the limited pedagogical
functions of contextualized conversations (Halliday and Matthiessen 2004). Our study
further confirms the limitation of this type of conversations in the interactional situation
of the participants having agreements on the topics.
In conclusion, this study presents the distributional patterns of conversations and
language expansions in the conversations in Chinese parent-child mealtime talk by
distinguishing parents’ and children’s contributions. The impact of the conversational
types on the distribution of language expansions was apparent. While the
decontextualized & conflicted conversations were the least occurring conversational
type, they had more language expansions, which created extensive language learning
26
opportunities for children in terms of exchanging information cohesively, hence being
higher quality of language that merits more attention by both parents and early
childhood educators. The most language expansions in these mealtime conversations
were extensions. This reflects the basic feature of these families’ mealtime
conversations characterized with the parents and the children’s additional ideas for
maintaining the conversations.
Limitations and implications
Before discussing the implications of the study, it is necessary to point out the
limitations of the study, which may be addressed in future research. The participating
families were all from well-educated professional backgrounds, thus do not cover all the
patterns of mealtime conversations in Chinese families from diverse SES. Future studies
should aim to collect mealtime conversations from Chinese families in different SES, so
a more complete picture of Chinese parent-child mealtime talk can be captured. Another
limitation concerned the video recording, which might affect the way the participants
talked as they were aware of being filmed. It could be possible that sometimes the
participants might not use the language which they normally use in more naturalistic
settings without video recording. Nonetheless, the videos recorded by the participants
themselves minimized the unnaturalistic effect than if the videos had been recorded by
researchers.
Despite these limitations, the findings of this study have important implications
as to how mealtime conversations could be better used as language learning occasions
for children. Firstly, the decontextualized & conflicted conversations seemed to create
more opportunities for the speakers to expand on each other’s ideas constantly. This
finding is consistent with the growing body of research that emphasize the didactic
value in decontextualized talk or conflicting conversations at the dinner table (Bohanek
27
et al 2009; Busch 2017; Snow and Beals 2006). Raising parents’ awareness of the
educational affordance of different conversational types, especially the value of
decontextualized & conflicted type is more likely to motivate them to adjust their
language for improving the conversational quality. Secondly, the Chinese parents’
initiation of around one fifth of conversations as contextualized & conflicted type raises
a concern about parents’ attitudes and practices relating to their children’s food intake
issues. This finding suggests that these parents may need some support to develop
constructive attitudes and strategies for encouraging children’s positive eating habits,
rather than using parental authority to enforce children to eat. Finally, although the
research was undertaken in the context of Chinese families, the finding of children’s
active contributions to the family dinner table conversations may have value that goes
beyond Chinese culture. Children’s participation and contribution to the mealtime
conversations creates a constructive language learning and practice environment for
themselves, hence, their participation should be encouraged by parents in order to
promote “learning about language” through spontaneous mutual conversations in daily
mealtime routines.
References
Arcidiacono, Francesco, and Antonio Bova. 2015. "Activity-Bound and Activity-
Unbound Arguments in Response to Parental Eat-Directives at Mealtimes:
Differences and Similarities in Children of 3-5 and 6-9 Years Old." Learning,
Culture and Social Interaction 6: 40-55. doi: 10.1016/j.lcsi.2015.03.002.
Antonio, Bova and Arcidiacono Francesco. 2014. "'You Must Eat the Salad Because It
Is Nutritious '. Argumentative Strategies Adopted by Parents and Children in Food-
28
Related Discussions at Mealtimes." Appetite 73: 81-94. doi:
10.1016/j.appet.2013.10.0.
Antonio, Bova, and Arcidiacono Francesco. 2013. "Investigating Children's Why-
Questions: A Study Comparing Argumentative and Explanatory Function."
Discourse Studies 15 (6): 713-34. doi: 10.1177/1461445613490013.
Antonio, Bova. 2019. "The Initial Phase of Argumentative Discussions between Parents
and Children." In the Functions of Parent-Child Argumentation, edited by
Antonio Bova, 39-63. Springer International Publishing. doi: 10.1007/978-3-
030-20457-03.
Aronsson, Karin, and Lucas GottAzén. 2011. "Generational Positions at a Family
Dinner: Food Morality and Social Order." Language in Society 40 (4): 405-26. doi:
10.1017/S0047404511000455.
Aukrust, Vibeke. (2002). ""What Did you Do in School Today?' Speech Genres and
Talk Ability in Multiparty Family Mealtime Conversations in Two Cultures." In
Talking to Adult: The Contribution of Multiparty Discourse to Language
Acquisition, edited by S. Blum-Kulda, & C. Snow, 55-83. Mahwah, NJ: Lawrence
Erlbaum.
Bohanek, Jennifer G., Robyn Fivush, Widaad Zaman, Caitlin E. Lepore, Shela
Merchant, and Marshall P. Duke. 2009. "Narrative Interaction in Family
Dinnertime Conversations." Merrill-Palmer Quarterly 55 (4): 488-515.
doi: 10.1353/mpq.0.0031.
Busch, Gillian. 2017. "'What Does It Say About It?': Doing Reading and Doing Writing
as Part of Family Mealtime." In Children’s Knowledge-In-Interaction, edited by
Amanda Bateman and Amelia Church, 297-311. Singapore: Springer. doi:
10.1007/978-981-10-1703-2_16.
29
Dai, Qian, Catherine McMahon, and Ai Keow Lim. 2020. "Cross-Cultural Comparison
of Maternal Mind-Mindedness among Australian and Chinese Mothers."
International Journal of Behavioral Development 44 (4): 365-70. doi:
10.1177/0165025419874133.
Demir, Özlem Ece, Meredith L. Rowe, Gabriella Heller, Susan Goldin-Meadow, and
Susan C. Levine. 2015. "Vocabulary, Syntax, and Narrative Development in
Typically Developing Children and Children with Early Unilateral Brain Injury:
Early Parental Talk about the 'There-and-Then' Matters." Developmental
Psychology 51 (2): 161-75. doi: 10.1037/a0038476.
Doan, Stacey N, and Qi Wang. 2010. "Maternal Discussions of Mental States and
Behaviors: Relations to Emotion Situation Knowledge in European American and
Immigrant Chinese Children." Child Development 81 (5): 1490-1503. doi:
10.1111/j.1467-8624.2010.01487.x.
Eggins, Suzanne. 2004. Introduction to Systemic Functional Linguistics. London:
Continuum International Publishing Group.
Ferdous, Hasan Shahid, Bernd Ploderer, Hilary Davis, Frank Vetere, and Kenton
O'Hara. 2016. "Commensality and the Social Use of Technology during Family
Mealtime." ACM Transactions on Computer-Human Interaction 23 (6): 1-26. doi:
10.1145/2994146.
Fiese, Barbara H, Kimberly P Foley, and Mary Spagnola. 2006. "Routine and Ritual
Elements in Family Mealtimes: Contexts for Child Well-Being and Family
Identity." New Direstions for Child and Adolescent Development. 2006(111), 67-
89. doi: 10.1002/cad.
30
Fruh, Sharon M., Jayne A. Fulkerson, Madhuri S. Mulekar, Lee Ann J. Kendrick, and
Clista Clanton. 2011. "The Surprising Benefits of the Family Meal." Journal for
Nurse Practitioners 7 (1): 18-22. doi: 10.1016/j.nurpra.2010.04.017.
Galatolo, Renata, and Letizia Caronia. 2018. "Morality at Dinnertime: The Sense of the
Other as a Practical Accomplishment in Family Interaction." Discourse and
Society 29 (1): 43-62. doi: 10.1177/0957926517726110.
Guo, Karen. 2015. "Motivation, Responsibility and Anxiety: Parenting Dispositions of
Chinese Mothers." Asia-Pacific Journal of Research In Early Childhood Education
10 (1): 1-19. doi: 10.17206/apjrece.2016.10.1.1.
Halliday, Michael Alexander Kirkwood. (1980) 2004. "Three Aspects of Children's
Language Development: Learning Language, Learning through Language,
Learning about Language. In The Language of Early Childhood (the collected
work of M. A. K. Halliday, Volume 4.), edited by Johnathan Webster, 308-326.
London: Continuum.
Halliday, Michael Alexander Kirkwood and Christian Matthiessen. 2004. An
Introduction to Functional Grammar. London: Hodder Arnold.
Hasan, Ruqaiya. 1996. "Literacy, Everyday Talk and Society." In Literacy in Society,
edited by Hasan Ruqaiya and Geoffrey Williams, 377-424. London: Longman.
Hu, Jiangbo, and Jane Torr. 2016. "A Study of Reasoning Talk between Australian
Chinese Mothers and Their Preschool Children : What Messages Are Mothers
Sending ?" Journal of Language, Identity & Education 15(3): 180-193. doi:
10.1080/15348458.2016.1169801.
Hu, Jiangbo, Jane Torr, Yonggang Wei, and Changhua Jiang. 2019. "Mealtime Talk as
a Language Learning Context : Australian Chinese Parents' Language Use in
Interactions with Their Preschool-Aged Children at the Dinner Table Mealtime."
31
Early Child Development and Care 191(3): 415-430. doi:
10.1080/03004430.2019.1621862.
Ji, Yan and Zhang Yiran. 2020. "An Investigation of Parent-Child Communication in
Families with Migrant Preschoolers." Early Childhood Education (27): 17-
20+29. doi:CNKI:SUN:YEJY.0.2020-27-004. [In Chinese].
Li, Xuan. 2020. "Fathers' Involvement in Chinese Societies: Increasing Presence,
Uneven Progress." Child Development Perspective 4(3):1-7. doi:
10.1111/cdep.12375.
Luo, Rufan, Catherine S., Tamis-LeMonda, and Lulu Song. 2013. "Chinese Parents'
Goals and Practices in Early Childhood." Early Childhood Research Quarterly 28
(4): 843-57. doi: 10.1016/j.ecresq.2013.08.001.
McHugh, Mary. 2012. "Interrater Reliability: The Kappa Statistic." Biochemia
Medica 22 (3): 276-282.
Rowe, Meredith. 2013. "Decontextualized Language Input and Preschoolers'
Vocabulary Development." Seminars in Speech and Language 34 (4): 260-266.
doi: 10.1055/s-0033-1353444.
Rydstrom, Helle. 2003. Embodying Morality: Growing up in Rural Northern Vietnam.
Honolulu, HI: University of Hawaii Press.
Snow, Catherine E, and Diane E Beals. 2006. "Mealtime Talk That Supports Literacy
Development." New Directions for Child and Adolescent Development 2006(111):
51-66. doi: 10.1002/cad.
Zhang, Heyi. 2018. "Relations of Parent-Child Interaction to Chinese Young Children’s
Emotion Understanding. " Joural of Pacific Rim Psychology 12(22): 1-9. doi:
10.1017/prp.2018.11.
