4.1 Semantic Prosody of Chansheng
4.1.2 Network Analysis
國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
SP and a secondary positive SP in the entire corpus, the positive SP tendency of chansheng seems to be invoked under the topics of entertainment, sports, and finance and the negative SP tendency was intensified under the topics of society, international and finance. On the other hand, the residuals also suggest that chansheng has a weaker negative SP tendency under the topic sports.
4.1.2 Network Analysis
The prototypical collocates under the topics that enhance either positive or negative SP tendency of chansheng were visualized from Figure 3 to Figure 8.
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Figure 3. Subset semantic network under topic society with negative collocates highlighted
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Figure 4. Subset semantic network under topic entertainment with positive collocates highlighted
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Figure 5. Subset semantic network under topic international with negative collocates highlighted
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Figure 6. Subset semantic network under topic sports with positive collocates highlighted
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Figure 7. Subset semantic network under topic finance with positive collocates highlighted
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Figure 8. Subset semantic network under topic finance with negative collocates highlighted These topic-dependent prototypical collocates of chansheng under different topics may fall into several semantic groups, as shown from Table 12 to Table 17. In these tables, only the top 20 positive or negative prototypical collocates were provided. However, in cases where the number of positive/negative collocates is lower than 20, all the prototypical positive/negative collocates (with degree of betweenness centrality greater than 0) were presented. To begin with topics that gave rise to the positive SP, we found that the positive collocates under topic
entertainment have semantic features of (i) acts and behavior (e.g., cushi ‘motivate’ and xiufu
‘fix/reconcile’), (ii) abstract concept (e.g., renzhi ‘cognition’), (iii) description about physical entity or abstract concept (e.g., rongyi ‘easy’ and mingxian ‘obvious’), and (iv) optimal change
‧
acts and behavior (e.g., tihui ‘experience’ and jiexiao ‘disclose’), (ii) abstract concept (e.g., shizhi ‘essense’ and xiaoyon ‘effectiveness’), (iii) optimal changes (e.g., cujin ‘promote’), (iv) prevention (e.g., yufan ‘prevent’) and (v) Attributes of physical entity/abstract concept (e.g., jida
‘very big’). The positive collocates within topic finance have semantic features of (i) acts and behavior (e.g., zijue ‘aware’), (ii) abstract concept (e.g., xiaoguo ‘effect’), (iii) attributes of physical entity/abstract concept (e.g., qimiao ‘amazing’), and (iv) balance (e.g., zhiheng ‘checks and balances’). These semantic features may be the contributor in increasing the positive SP tendency under the three topics.
In terms of topics that enhanced the negative SP tendency, we observed that the negative collocates under topic society have semantic features of (i) unpleasant emotion, thought and mental states (e.g., jaolu ‘anxiety’ and kongju ‘fear’), (ii) acts and behavior (fanyin ‘reaction’), (iii) health issues (e.g., bingbian ‘pathological changes’ and fayan ‘inflammation’), (iv)
destructive/harmful/annoying events (e.g., suanshan ‘damage/injure’), and (v) change of state (e.g., jiandi ‘lower’). The negative collocates under topic international were found to have semantic features of (i) unpleasant emotion, thought and mental states (e.g., jaolu ‘anxiety’ and yuanhen ‘hatred’), (ii) acts and behavior (e.g., yaohuang ‘swaying’), (iii) health issues (e.g., guomin ‘allergic’), (iv) change of state (e.g., biancha ‘getting worse’), and (iv) natural
phenomenon (e.g., yanwu ‘smoke’). Lastly, the semantic features of negative collocates for topic finance are (i) unpleasant emotion, thought and mental states (e.g., wuhui ‘misunderstanding’
and maoduen ‘paradox’), (ii) acts and behavior (e.g., yapo ‘oppress’), (iii) bad consequence (e.g., fuzuoyong ‘side-effect’), (iv) undesirable description about physical entity/abstract concept (e.g., buliang ‘not-well’), and (v) loss (suenshi ‘loss’ and luocha ‘undesirable difference’). Also, it is
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
possible that these semantic features are the source that intensify the negative SP tendency under the three topics.
Taken together, we found that the shared semantic features in topics where the positive SP tendency was increased are (i) acts and behavior and (ii) abstract concept. For the topics that intensified the negative SP tendency, the common semantic features are: (i) unpleasant emotion, thought and other mental states, (ii) acts and behavior, and (iii) health issues.
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Negative prototypical collocates under topic society Semantic features Prototypical collocates
Topic Society ✕ negative SP Unpleasant emotion, thought
and mental states
焦慮jaolu ‘anxiety’, 負面 fumian ‘negative’, 恐懼 kongju
‘fear’, 壓力 yali ‘pressure’, 誤解 wujie ‘misunderstanding’,
敏感 minggan ‘sensitive’, 疑慮 yilu ‘doubt’
Acts and behavior 反應 fanyin ‘reaction’, 干擾 ganrao ‘disturb’
Health issues 惡化 ehua ‘deteriorate’, 病變 bingbian ‘pathological
changes’, 發炎 fayan ‘inflammation’, 中毒 zhongdu
‘poisoned’, 不適 bushi ‘discomfort’
Destructive/harmful/annoying events
損傷 suanshan ‘damage/injure’, 引爆 yinbao ‘detonate’, 爆炸
baozha ‘explode’, 噪音 zaoyin ‘noise’
Change of state 降低 jiandi ‘lower’
Others 強烈 qianglie ‘intense’
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Table 13
Top 20 positive prototypical collocates under topic entertainment Semantic features Prototypical collocates
Topic Entertainment ✕ positive SP
Acts and behavior 促使 cushi ‘motivate’, 修復 xoufu ‘fix/reconcile’, 舒緩 shuhuan
‘smooth’, 信任 xinren ‘trust’
Abstract concept 認知 renzhi ‘cognition’, 效果 xiaoguo ‘effect’, 信心 xinxin
‘faith’
Description about physical entity or abstract concept
容易 ronyi ‘easy’, 明顯 mingxian ‘obvious’,極大 jida ‘very-big’,
重大 zhongda ‘significant’, 濃厚 nonghou ‘rich’
Optimal change 改善 gaishan ‘improve’, 增長 zengzhang ‘increase’
Others 進而 jiner ‘so-as-to’, 造成 zaocheng ‘bring-about’, 有助 youzhu
‘helpful’, 將會 jianhui ‘will-do’, 光澤 guanze ‘luster’
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Top 20 negative prototypical collocates under topic international Semantic features Prototypical collocates
Topic International ✕ negative SP Unpleasant emotion,
thought and mental states
焦慮jaolu ‘anxiety’, 負面影響 fumianyingxiang ‘negative impact’, 誤解 wujie ‘misunderstand’, 疑慮 yilu ‘doubt’, 怨恨
yuanhen ‘hatred’, 衝擊 chongji ‘shock’, 自卑 zibei ‘self-abased’
Acts and behavior 摩擦 moca ‘friction‘, 搖晃 yaohuang ‘swaying’, 衝動
chongdong ‘impulsive’
Health issues 副作用 fuzuoyong ‘side-effect‘, 過敏 guomin ‘allergic’,後遺症
hoyizheng ‘sequela’, 病變 bingbian ‘pathological changes’, 暈
眩 yunxuan ‘dizzy’
Change of state 變差 biancha ‘getting worse’, 變化 bianhua ‘change’, 減低
jiandi ‘reduce’,
Natural phenomenon 煙霧 yanwu ‘smoke‘, 燃燒 ranshao ‘burning’
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Table 15
Top 20 positive prototypical collocates under topic sports Semantic features Prototypical collocates
Topic Sports ✕ positive SP
Acts and behavior 體會 tihui ‘experience’, 揭曉 jiexiao ‘disclose’, 吸收 xishou
‘absorb’, 決議 jueyi ‘decide’, 提名 timing ‘nominate’
Abstract concept 實質 shizhi ‘essense’, 效用 xiaoyon ‘effectiveness’, 效果
xiaoguo ‘effect’, 動機 dongji ‘motivation’, 商機 shanji
‘business opportunity’, 成就感 chengjiougan ‘sense-of-achivement’
Optimal change 促進 cujin ‘promote’, 新增 xinzeng ‘add’
Prevention 預防 yufan ‘prevent’
Attributes of physical entity/abstract concept
極大 jida ‘very big’, 常見 chanjian ‘often-seen’, 獨特 dute
‘unique’, 自然 ziran ‘natural’
Others 進而 jiner ‘so-as-to’ , 現行 xiangxing ‘be-in-effect’
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Positive prototypical collocates under topic finance Semantic features Prototypical collocates
Topic Finance ✕ positive SP Acts and behavior 自覺 zijue ‘aware‘
Abstract concept 效果 xiaoguo ‘effect’, 效用 xiaoyon ‘effectiveness’
Attributes of physical entity/abstract concept
極大 jida ‘very big’, 奇妙 qimiao ‘amazing’, 巨大 juda
‘gigantic’, 微妙 weimiao ‘mystical’
Balance 制衡 zhiheng ‘checks and balances’, 均勻 junyun ‘balance’
Others 進而 jiner ‘so-as-to’, 必將 bijian ‘inevitably’, 共鳴 gongming
‘resonance’
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Table 17
Top 20 negative prototypical collocates under topic finance
Semantic features Prototypical collocates
Topic Finance ✕ negative SP Unpleasant emotion, thought
and mental states
疑慮 yilu ‘doubt’, 誤會 wuhui ‘misunderstanding’, 憂鬱 yolu
‘melancholic’, 矛盾 maoduen ‘paradox’, 疑惑 yihuo ‘doubt’,
罪惡感 zuiegan‘guilty’, 負面 fumian ‘negative’, 負面影響 fumianyingxiang ‘negative impact’
Acts and behavior 對立 duili ‘oppose’,摩擦 moca ‘friction‘,危害 weihai ‘harm’,
壓迫 yapo ‘oppress’, 扭曲 niuqu ‘distort’
Bad consequence 副作用 fuzuoyong ‘side-effect‘, 後遺症 hoyizheng ‘sequela’
Undesirable description about physical entity/abstract concept
不良 buliang ‘not-well’
Loss 損失 suenshi ‘loss’, 落差 luocha ‘undesirable difference‘
Others 細菌 xijun ‘bacteria‘,臭味 chowei ‘smelly odor’
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
4.2.1 Concordance Semantic Prosody Analysis
Figure 9 shows the overall distribution of the evaluative meanings of the concordances of the node word niangcheng under the ADN corpus. The concordances with negative evaluation account for about 74.5 %, positive evaluation 22.1 and neutral evaluation sentiments 3.4 % of all the data. The node word niangcheng has a strong negative SP.
Figure 9. Distribution of the evaluative meanings of the concordances of the node word niangcheng under the seven topics
‧
Table 18 shows the SP distribution of the concordances of the node word niangcheng across the seven topics. A chi-square test shows that there is a significant association between
TOPIC and SEMANTIC PROSODY (χ2 =128.17, df=12, p= p<.001, Cramer’s V=0.15).
Table 18
Standardized Residuals in a Chi-Square Contingency Table for the SP distribution of the node word niangcheng under each topic
χ2=128.17, df=12, p<.001, Cramer’s V=0.15
Society Politics Enter. Inter. Sports Finance Lifestyle
Positive Observed
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
meaning are much more than expected under the topics of sports and lifestyle, and the proportion of negative evaluative meaning is much more than expected under topic
international. With the strong negative SP, the positive SP tendency of niangcheng is elicited under the topics of sports and lifestyle and negative SP tendency is further intensified under topic international.
4.2.2 Network Analysis
Again, the prototypical collocates under topics that enhance either positive or negative SP tendency of niangcheng are visualized from Figure 10 to Figure 12.
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Figure 10. Subset semantic network under topic international with negative collocates highlighted
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Figure 11. Subset semantic network under topic sports with positive collocates highlighted
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Figure 12. Subset semantic network under topic lifestyle with positive collocates highlighted According to semantic features, the prototypical collocates of niangcheng under these topics are classified into different semantic subsets as shown in from Table 19 to Table 21. In topics where the positive SP tendency was elicited, the semantic features of the positive
collocates for topic sports are (i) desirable description (e.g., xinghao ‘fortunate’) and (ii) settling issues. For topic lifestyle, the positive collocates have semantic features of (i) desirable
description (e.g., ronyi ‘easy’ and zhongda ‘significant’) and (ii) medical manner (e.g., zhiliao
‘treatment’). These semantic features may be the source that triggered the positive SP tendency.
The only shared semantic feature in the two topics is desirable description.
‧
international, the semantic features of the negative collocates are (i) acts and behavior (e.g., tuiji
‘push’), (ii) destructive/harmful events, accidents and disaster (e.g., zainan ‘disaster’ and baozha
‘explode’) and (iii) natural disaster (e.g., haixiao ‘tsunami’). The enhancement of negative SP tendency may due to these semantic features. Noticeably, most of the negative collocates are associated with the second semantic set.
Table 19
Top 20 negative prototypical collocates under topic international Semantic features Prototypical collocates
Topic International ✕ negative SP Acts and behavior 推擠 tuiji ‘push’
Destructive/harmful events, accidents and disaster
意外 yiwai ‘accident’, 災難 zainan ‘disaster’, 傷亡 shanwang
‘casualties’, 事故 shigu ‘incident’, 車禍 chehuo ‘car accident’, 火 災 huozai ‘fire disaster’, 爆炸 baozha ‘explode’, 失控 shikong
‘out of control’, 縱火 zonghuo ‘arson’, 災害 zaihai ‘disaster’, 喪生 sangsheng ‘die’, 動亂 dongran ‘turmoil’, 起火 qihuo ‘outbreak of fire’, 死傷 sishang ‘death and injury’, 喪命 sangming ‘die’, 衝突 chongtu ‘conflict’, 悲劇 beijui ‘tragedy’
Natural disaster 海嘯 haixiao ‘tsunami’
Others 不慎 bushen ‘accidentally’
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
Table 20
Positive prototypical collocates under topic sports Semantic features Prototypical collocates
Topic Sports ✕ positive SP
Desirable description 幸好xinghao ‘fortunate’, 還好 haihao ‘fortunate’, 及時 jishi
‘on-time’
Settling issues 化解 huajie ‘resolve’
Table 21
Positive prototypical collocates under topic lifestyle
Semantic features Prototypical collocates
Topic Lifestyle ✕ positive SP
Desirable description 容易 ronyi ‘easy’, 重大 zhongda ‘significant’, 最好 zuihao ‘the best’, 所幸 suoxing ‘fortunate’, 可能 keneng ‘possible’
Medical manner 治療 zhiliao ‘treatment’
Others 造成 zaocheng ‘bring-about’
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
4.3.1 Concordance Semantic Prosody Analysis
Figure 13 shows the overall distribution of the concordances of the node word cucheng under the ADN corpus. The concordances with positive evaluation accounts for 74.2 %, neutral evaluation 5.8 %, and negative evaluation 19.9 % of the data. The node word cucheng shows a strong positive SP.
Figure 13. Distribution of the evaluative meanings of the collocates of the node word cucheng in the ADN
‧
Table 22 shows the SP distribution of the concordances of the node word cucheng across the seven different topics. A chi-square test shows that there is a significant association between
TOPIC and SEMANTIC PROSODY (χ2 =54.93, df=12, p <.001, Cramer’s V=0.1).
Table 22
Standardized Residuals in a Chi-Square Contingency Table for SP distribution the node word cucheng under each topic
χ2=54.93, df=12, p<.001, Cramer’s V=0.1
Society Politics Enter. Inter. Sports Finance Lifestyle
Positive Observed
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
meaning is much more than expected under topic international. This suggests that the negative SP tendency of cucheng is triggered under topic international despite its strong positive SP nature.
4.3.2 Network Analysis
The negative prototypical collocates of cucheng under topic international that triggers the negative SP tendency of the node word is presented in Figure 14.
Figure 14. Subset semantic network under topic international with negative collocates highlighted
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
The topic-dependent negative prototypical collocates under topic international can be generalized as (i) unpleasant emotion, thought and other mental states (e.g., maoduen ‘paradox’), (ii) termination of events (e.g., ezhi ‘stop’), (iii) warfare (e.g., lenzhan ‘cold war’ and jiangju
‘deadlock’), and (iv) change of state (e.g., zhuanbian ‘change’) (see Table 23). These semantic features may account for the elicitation of the negative SP tendency.
Table 23
Top 10 negative prototypical collocates under topic international Semantic features Prototypical collocates
Topic International ✕ negative SP Unpleasant emotion,
thought and other mental states
矛盾maoduen ‘paradox’
Termination of events 終結 zhongjie ‘terminate’, 遏止 ezhi ‘stop‘, 落幕 luomu ‘ending’
Warfare 冷戰 lenzhan ‘cold war’, 僵局 jiangju ‘deadlock’, 敵對 didui
‘hostility’, 派系 paixi ‘faction’
Change of state 轉變 zhuanbian ‘change’, 削減 xiaojian ‘reduce’
4.4 Internal Summary
Throughout the entire corpus, the SP distribution of the concordance lines for each node was summarized in Table 24. The node word chansheng has a primary negative SP with a secondary
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
respectively.
Table 24
Semantic prosody profile of the three node words
Node word Positive % Neutral % Negative % Semantic prosody
chansheng 29.1 10.6 60.3 Negative
niangcheng 22.1 3.4 74.5 Negative
cucheng 74.2 5.8 19.9 Positive
The topic-dependent SP distribution of the concordance lines of the three node words was summarized in Table 25. In the diagram, “positive” indicates a positive SP, and “negative” refers to a negative SP. A red label with an upward-pointing arrow indicates the SP distribution has a residual value greater than 1.96 in a topic; conversely, a green label with a down-facing arrow indicates the SP distribution has a residual value lower than -1.96 in a topic. Two important patterns should be noted. First, a consistent increment of the proportion of the negative SP was observed under topic international. Secondly, the arousal of positive SP of chansheng and niangcheng was discovered under topic sports.
‧
Comprehensive view of SP tendencies across topics
Society Politics Enter. Inter. Sports Finance Lifestyle
產生
In terms of the semantic features of the prototypical collocates (“prototypical features”
for convenience) of a node word under topics that intensify the positive/negative SP tendency of that node word, some phenomena were detected. The common positive prototypical features for chansheng were (i) acts and behavior and (ii) abstract concept; for niangcheng was desirable description. On the other hand, the common negative ones for chansheng were (i) unpleasant emotion, thought and other mental states, (ii) acts and behavior, and (iii) health issues; for niangcheng was destructive/harmful events, accidents and disaster.
‧
The current study has shown that (1) chansheng has a primary negative SP and a secondary positive SP, (2) niangcheng a strong negative SP, and (3) cucheng a strong positive SP in the ADN corpus. In parallel to Xiao and McEnery (2006), the last two node words echo the findings in their study; nonetheless, chansheng contradicts to their finding where the node word was found to have a mixed SP (31% negative and 24 % positive). A possible reason for the surge of negative SP is the influence of negative disposition under news genre. In journalism, it is argued that “the more negative the event in its consequences, the more probable that it will become a news item” (Cohen & Young, 1981, p. 56). Thus, the negative SP tendency of chansheng, which has a mixed SP in domain-general texts (hereafter ‘general’), may be amplified under the ADN corpus. Compatible to the SP trend of news, niangcheng, which has a negative SP in general, continues to show a strong negative SP. On the other hand, despite exposing to the overall sentimental orientation of news genre, cucheng nonetheless carries a strong positive SP. Its strong positive SP in general may entail its insensitivity to the topical influence of the news genre.
The SP tendency of the mixed-SP node word, chansheng, varies to a great degree across different topics; conversely, the SP tendencies of the strong-SP node words, niangcheng and cucheng, are relatively stable despite exposing to diverse topic types. As presented in the
individual result section of each node word, the Cramer’s V value informs that only the strength of the association between TOPIC and SEMANTIC PROSODY is moderately strong for chansheng, but weak for the other two node words. Thus, we suggest that a mixed-SP node word in general is susceptible to the topic change in a way that it serves as a vehicle to resonate the overall sentiment and content nature of a given topic. On the other hand, a strong-SP node word in
‧
general is rather insensitive to the topic change and has a more monotonous SP usage. Since only the mixed-SP node word is found to be influenced by topic, the moderate hypothesis is
confirmed in our study. The weak influence of topic on the strong-SP node words rejects the strong hypothesis. Lastly, the null hypothesis of no correlation existing between topic and the SPs of any node words is not supported by the evidence of topic-dependent SP of chansheng.
Along with the well-validated notion of register prosody by O'Halloran (2007), we further propose the idea of topic prosody—context-dependent SP operates at the level of topic.
Based on the findings in the present study, the phenomenon of a topic-dependent SP is confined to chansheng, the mixed-SP node word. We might also argue that topic has higher probability in selecting the SP phenomenon of mixed-SP node words more often than strong-SP node words.
In semantic network analysis, topic-dependent prototypical features generalized from prototypical collocates of the three target node words may contribute to the increase of
positive/negative SP tendency for these node words. To find out how the prototypical features are connected to the main themes of the news article under a given topic, we decided to utilize the latent Dirichlet allocation (LDA) algorithm (Blei, Ng, & Jordan, 2003) to explore the typical semantic content/matter of the news coverage under each topic. LDA is an effective data-mining technique which offers analysts a quick look of the thematic distribution across documents under a collection of texts. Moreover, LDA has been employed for tasks such as topic discovery in scientific papers (Griffiths & Steyvers, 2004) and recommendation for data annotation (Krestel, Fankhauser, & Nejdl, 2009). We applied LDA to the six topics where the SP of one of the node words is intensified significantly. We selected an arbitrary number and classified the news collection of each topic into ten themes. The naming for each theme, along with topic 15 relevant
‧ 國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
finance, lifestyle) are shown from Appendix C to Appendix H.
Listed below are themes generalized by LDA within the aforementioned topics; these themes contain the news subject matter that journalists typically report about. Among them, we provide a few concordances demonstrating how the prototypical semantic features of a node word relate to the content of news articles.
5.1 Topics Triggering Positive SP
Under topic entertainment, it is observed that the positive tendency of chansheng rises. The typical ten themes are jingpin ‘boutique’, biaoyan ‘performance’, falujiufen ‘lawsuit’, baoyang
‘health maintenance’, shenghuoyule ‘life entertainment’, yiliao ‘medical care’, ganqingbagua
‘gossiping’, yingshi ‘television’, jiating ‘family’, and dianying ‘movie’ (see Appendix D).
Examples expressing positive evaluation for the topic are provided in (7).
Examples expressing positive evaluation for the topic are provided in (7).