Abstract
This study conducts a corpus-based diachronic investigation of the noun phrase hóngyǎn (红眼) and proposes the diachronic cognitive pragmatic model to demonstrate that particularized conversational implicatures develop into generalized conversational implicatures and eventually become new coded meanings. The results show that (1) different types of meanings such as generalized conversational implicatures and particularized conversational implicatures are closely correlated with each other; (2) the diachronic change from particularized conversational implicatures to new coded meanings through generalized conversational implicatures is an entrenchment-conventionalization process; (3) conventions and contexts coexist with each other, and meaning change is from highly contextualized to highly conventionalized. This study reveals that pragmatic meanings can become entrenched and conventionalized to be new coded meanings through repeated usage and suggests that the diachronic corpus method contributes to uncovering the connection between different types of meanings.
Similar content being viewed by others
Introduction
Grice (1989) defines conversational implicature as the meaning implied by the speaker, which is derived from the Cooperative Principle and its four conversational maxims, such as Quality, Quantity, Relation, and Manner. He further divides conversational implicatures into generalized conversational implicatures (hereafter GCIs) and particularized conversational implicatures (hereafter PCIs). The former are the implicatures closely associated with the use of a certain form of words in the absence of special contexts, with the assumption that the speaker observes conversational maxims, while the latter are the context-dependent implicatures in the sense that the speaker violates conversational maxims (Grice, 1989). For example,
(1) John: How did yesterday’s lecture go?
Mary: Some students left before it ended.
+> a. Not all students left before the lecture ended.
+> b. The lecture didn’t go well.
In (1), “some students left before it ended” implicates that “not all students left before the lecture ended,” which, in Grice’s sense, is considered as a GCI, and this reading can be explained by observing conversational maxims in the absence of the special context. By contrast, “some students left before it ended” has a PCI, i.e., the lecture didn’t go well, because this utterance violates the Relation maxim. It is indicated in (1) that there is a clear divide between GCIs and PCIs.
Grice’s GCI-PCI divide provokes a heated debate in post-Gricean pragmaticsFootnote 1. Some scholars propose different meaning distinctions from a synchronic perspective (Bach, 1994; Carston, 2002, 2019, 2021; Levinson, 2000, 2024; Sperber and Wilson, 1995), while other scholars argue that there exists a diachronic change of different types of meanings (Traugott, 2012, 2018, 2019; Traugott and Dasher, 2002). Ehmer and Rosemeyer (2018, p. 541) point out, “historical linguists began exploiting the notion of implicature in the description of meaning change.” On this basis, this study is intended to prove that PCIs develop into GCIs and finally become new coded meanings, arguing that there is a diachronic change from PCIs to GCIs.
The paper is organized as follows. The introduction part is followed by the section “Literature review.” In the section “Analytical framework and research methodology,” the diachronic cognitive pragmatic model and the adjusted frequency measure are delineated. Section “Ad-hoc inferences of PCIs of hóngyǎn” elaborates on ad-hoc inferences of PCIs of hóngyǎn in ancient Chinese, modern Chinese, and contemporary Chinese, respectively. In the section “The entrenchment process of PCIs of hóngyǎn,” we investigate the entrenchment process. Section “The conventionalization process of GCIs of hóngyǎn” analyzes the conventionalization process. Finally, some concluding remarks are provided in the section “Conclusion.”
Literature review
“Since its inception, Grice’s classical theory of conversational implicatures has revolutionized pragmatic theorizing, giving rise to a large number of reinterpretations, revisions, and reconstructions” (Huang, 2014, p. 43). Currently, within the philosophy of language and pragmatics, two approaches can be identified in post-Gricean pragmatics: the conversational-implicature approach and the non-conversational-implicature approach, in an endeavor to settle the question in Grice’s dichotomy (Huang, 2017a). In light of the conversational-implicature approach, Levinson (2000, 2024) distinguishes utterance-type meanings from utterance-token meanings. Utterance-type meanings, which are almost the same as GCIs, are “matters of preferred interpretations—the presumptive meanings—which are carried by the structure of utterances, given the structure of the language, and not by virtue of the particular contexts of utterance” (Levinson, 2000, p. 1). Compared with context-independent utterance-type meanings, utterance-token meanings, which are akin to PCIs, are the novel, nonce, or once-off use of language. According to Levinson (2000), utterance-type meanings are governed by Q-, I-, and M- principlesFootnote 2, which are triggered by conventional use of language.
(2) John took out the key and the door opened.
+> John took out the key and then the door opened.
In (2), by I-principle, the use of a semantically general expression “and” I-implicates a semantically specific interpretation “and then.” Here, “and” conveys the utterance-type meaning “and then.”
Different from the conversational-implicature approach, the non-conversational-implicature approach is mainly represented by Relevance Theorists and Bach. Relevance Theorists (Carston, 2002; Sperber and Wilson, 1995) cut across the GCI-PCI divide and put forward the explicature/implicature distinction instead. Carston (2009, p. 47) points out, “a proposition communicated by an utterance U is explicit if and only if it is a development of a logical formFootnote 3 encoded by U; any other communicated proposition is an implicature.” In view of this, explicatures (GCIs in Grice’s sense) are developed from logical forms encoded by the utterance, while implicatures (PCIs in Grice’s sense) are not related to logical forms (Carston, 2009). Since Relevance Theory insists on the linguistic underdeterminacy thesis, i.e., the linguistically encoded meaning of a sentence underdetermines the intentional meaning expressed by the speaker using the sentence, and it must be determined in the context (Carston, 2002), explicatures and implicatures both have strong context-sensitivity under the guidance of the principle of RelevanceFootnote 4.
(3) Max: Would you like some supper?
Amy: I’ve eaten.
+> a. I’ve eaten [a meal this evening].
+> b. Amy is declining Max’s offer of some supper.
In (3), “I’ve eaten” is an incomplete logical form, which needs to be explicitly completed, yielding an explicature “I’ve eaten [a meal this evening],” and “I’ve eaten” is also an implicitly communicated proposition, conveying that “Amy is declining Max’s offer of some supper,” an implicature in the relevance-theoretic model. Relevance Theorists (Carston, 2002; Wilson and Sperber, 2004) reject the role of conventions in the inference of explicatures and implicatures.
In agreement with Relevance Theorists, Bach (1994) abandons Grice’s GCI-PCI distinction, putting forward the impliciture/implicature distinction. Implicitures, which are like Grice’s GCIs, go beyond what is said but are built out of what is said, while implicatures, which are equal to Grice’s PCIs, are the additional meanings external to what is said (Bach, 1994). For example,
(4) Jim: How was the party? Did it go well?
Sue: There wasn’t enough alcohol and everyone left early.
+> a. There wasn’t enough alcohol and everyone [in the party] left early.
+> b. The party was not a great success.
In (4a), “everyone” needs to be enriched into “everyone in the party,” fleshing out the proposition and yielding an impliciture “there wasn’t enough alcohol and everyone [in the party] left early.” By comparison, (4b) is an implicature, which implies that “the party was not a great success.” Implicitures are governed by Take-for-granted and Not-worth-considering principlesFootnote 5. Bach (1998) adopts the term “standardization,” instead of conventionalization, to account for context-dependent implicitures. Standardization means that “a form of words is standardized for a certain use if this use, though regularized, goes beyond literal meaning and yet can be explained without special conventions” (Bach, 1998, p. 712).
Based on the above review of the conversational-implicature approach and the non-conversational-implicature approach, different views on meaning distinctions in post-Gricean pragmatics are summarized in Table 1.
As shown in Table 1, there are three issues that need to be addressed in this study. The first issue is about the relationship between different types of meanings, such as GCIs and PCIs. Post-Griceans (Bach, 1994; Carston, 2002; Levinson, 2000; Sperber and Wilson, 1995) propose new labels for different meaning distinctions to highlight Grice’s GCI-PCI divide. They mainly rely on synchronic research to discuss meanings, ignoring the diachronic change of meanings (Andersen, 2015). The relationship between different types of meanings has not been well investigated from the perspective of diachronic change in post-Gricean pragmatics. Post-Griceans argue that there is a divide between different types of meanings, and then break up the relationship between them, which has raised doubts among historical linguists about the validity of meaning distinctions such as the GCI-PCI distinction. In historical linguistics, diachronic semantics is an important field gaining in popularity (Deo, 2015). There is a diachronic change of different types of meanings (Ehmer and Rosemeyer, 2018; Traugott, 2012, 2018, 2019; Traugott and Dasher, 2002). However, post-Griceans largely neglect meaning change. Therefore, the relationship between different types of meanings still needs to be further explored from a diachronic perspective.
The second issue is about the derivation of conversational implicatures. Post-Griceans (Bach, 1994; Carston, 2002; Levinson, 2000; Sperber and Wilson, 1995) put forward different pragmatic principles for the derivation of conversational implicatures. But, Grice (2001) in Aspects of Reason argues that not all conversational implicatures rely on pragmatic principles such as the Cooperative Principle when distinguishing between two ways of reasoning, i.e., the hard way and the quick way. The hard way of reasoning is a step-by-step process in which the hearer consciously and laboriously infers the speaker’s intentional meaning by pragmatic principles, while the quick way of reasoning is a shortcut process in which the hearer unconsciously and effortlessly jumps to conclusions and infers what the speaker intends to express by habituation (Zhang and Zhang, 2015). In other words, the hard way emphasizes pragmatic principles, while the quick way highlights habituation. However, post-GriceansFootnote 6 hold that working out conversational implicatures relies on pragmatic principles, largely neglecting the quick way of reasoning.
The third issue is that the relationship between conventions and contexts still needs to be further explored. Post-Griceans seem to suggest that conventions and contexts are mutually exclusive. For instance, Levinson (2000) argues that utterance-type meanings are triggered by conventions of language use, independent of contexts. Relevance Theorists (Carston, 2002; Wilson and Sperber, 2004) deny the role of conventions, intensifying the role of contexts. They (Kolaiti and Wilson, 2014; Wilson and Kolaiti, 2017) take the noun phrase red eyes as an example, extract the data from The Bank of English, and statistically analyze the occurrences of the enriched meanings of red eyes in different contexts. The results show that among 26 discourse contexts in which red eyes appears, it expresses different enriched meanings in 17 contexts, which demonstrates that the enriched meanings are highly context-dependent. In fact, Relevance Theory also realizes that some metaphorical inferences gradually become conventionalized over time, so that the hearer can speed up the steps in the inferential process and get the meaning directly (Vega Moreno, 2007). Both conventions and contexts occupy a part in meaning generation.
As a matter of fact, Grice (1989, p. 39) points out, “it may not be impossible for what starts life, so to speak, as a conversational implicature to become conventionalized.” In Grice’s view, “meaning change results from the conventionalization of conversational implicatures” (Ehmer and Rosemeyer, 2018, p. 541). In order to further discuss the above three issues, this study investigates the diachronic change of conversational implicatures from a diachronic cognitive pragmatic approach.
Analytical framework and research methodology
Analytical framework
In this study, the diachronic cognitive pragmatic model (hereafter DCPM), as the analytical framework, has two theoretical sources. The first theoretical source is historical pragmaticsFootnote 7, a usage-based approach to meaning change (Traugott, 2004). Following Levinson, i.e., “it is possible to argue that there is a sequence from particularized through generalized conversational implicatures to conventional implicatures” (Levinson, 1979, p. 216)Footnote 8, Traugott (2012, 2018, 2019) and Traugott and Dasher (2002) propose the Invited Inferencing Theory of Semantic Change (hereafter IITSC) to characterize the diachronic change of language use. According to IITSC, there is a path from coded meanings to utterance-token meanings to utterance-type meanings to new coded meanings. Here, utterance-token meanings and utterance-type meanings refer to PCIs and GCIs, respectively (Levinson, 2000; Traugott and Dasher, 2002). For example, Traugott and Dasher (2002) conduct a study of as/so long as to investigate its development. The results showed that in Old English and Middle English, the spatial meaning coexisted with the temporal “for the same length of time as”; in Early Modern English, the conditional PCI “provided that” was generalized to contexts in which the conditional was more salient, and at this stage the conditional meaning became a GCI of temporal as/so long as; by the mid-nineteenth century the conditional appeared to be the only possible meaning, and the GCI was semanticized as a conditional polysemy of temporal as/so long as.
In IITSC, the individual hearer derives utterance-token meanings in different contexts via invited inferencingFootnote 9. As utterance-token meanings become more salient in a community, they are gradually conventionalized among a community as utterance-type meanings (Traugott, 2018). And utterance-type meanings become conventionalized/semanticized as new coded meanings. According to Traugott and Dasher (2002), utterance-token meanings are stored in individual minds, and utterance-type meanings are stored in collective minds. This view of meaning change reflects the shift from the individual to the collective. In a nutshell, pragmatic meanings are eventually conventionalized into new coded meanings.
The second theoretical source is Schmid’s Entrenchment-and-Conventionalization Model (hereafter EC-Model). As a usage-based approach, the EC-Model centers on the entrenchment-conventionalization process of meaning change from individual cognition to collective society (Schmid, 2014, 2015, 2016, 2020, 2023). Language becomes entrenched and conventionalized through repeated usage activities in usage events. Entrenchment is “the continual reorganization of linguistic knowledge in the minds of speakers, which is driven by repeated usage activities in usage events” (Schmid, 2020, p. 2). It essentially refers to the routinization and schematization of associations. Conventionalization is “the continual process of establishing and readapting regularities of communicative behavior among the members of a speech community, which is achieved by repeated usage activities in usage events and subject to the exigencies of the entrenchment processes taking place in the minds of speakers” (Schmid, 2020, p. 2). It involves the usualization and diffusion of utterance typesFootnote 10. In short, entrenchment is the cognitive process operating in the minds of individual speakers, and conventionalization is the social process taking place in a speech community (Schmid, 2020; Verspoor and Schmid, 2024).
The EC-Model takes the view that language is constantly generated and updated through the interaction between entrenchment and conventionalization. Frequency of repeated usage plays a large part in the interaction between entrenchment and conventionalization, because entrenched language use in individual minds gradually forms the contract obeyed by all the members of a speech community through repeated usage (Schmid, 2015, 2020). Hence, entrenched language use is eventually conventionalized into the collective minds of all the members of a speech community. Repeated usage finally becomes fixed language expressions (Schmid, 2015). Fixed language expressions can only stimulate the expected response in the minds of the hearer when their use characteristics are conventionalized.
As mentioned above, Traugott’s historical pragmatics and Schmid’s EC-Model have the following three similarities. First, they both base meaning change on the usage-based view. Second, they highlight that conventionalization plays a significant role in the process of meaning change. Third, meanings develop from the individual to the collective. These three similarities determine a strong integration of historical pragmatics and the EC-Model. Currently, post-Gricean pragmatics mainly adopts the synchronic method and strictly distinguishes different types of meanings, ignoring the diachronic change of meanings. Different from post-Gricean pragmatics, this study constructs a new model to account for meaning change, namely DCPM (see Fig. 1).
The diachronic cognitive pragmatic model.
Figure 1 presents a sketch of DCPM, which explores the entrenchment-conventionalization process of meaning change from a diachronic perspective. This model mainly has the following four aspects. First, PCIs are derived by virtue of ad-hoc inferences which refer to the product of the interaction between the individual speaker and the individual hearer in a speech event and involve joint construction of meanings. Ad-hoc inferencesFootnote 11 may comprise metonymization, metaphorization, and metaphtonymization in meaning change. In the process of ad-hoc inferences, PCIs of linguistic expressions are extended through the processes of metonymization, metaphorization, and metaphtonymization in specific contexts. The extension of word meanings mainly depends on the cognitive thinking structure of metonymy, metaphor, and metaphtonymy.
Metonymization refers to the formation of metonymy, which is a mapping within a cognitive domain, i.e., the intra-domain mapping. For example, in the sentence Our team needs some new heads, “new heads” refer to “new people,” which belongs to the same cognitive domain, that is, the human-body domain, going through metonymization. Metaphorization means the formation of a metaphor, which is the mapping between two cognitive domains, i.e., the inter-domain mapping from the source domain to the target domain (Geeraerts, 2010). For example, in the sentence time is money, it covers two different cognitive domains, mapping from the money domain to the time domain, which is metaphorical. With respect to metaphtonymization, it is used to represent the formation of metaphtonymy, involving the interaction between metaphor and metonymy (Goossens, 2003). For example, in the expression catch the minister’s ear, Goossens (2003) argues that the minister’s ear can be metaphorically interpreted as an entity on the move, and it is metonymically used for the minister and for the minister’s attention.
Second, entrenchment is a process of individual cognition. PCIs become entrenched into GCIs through repeated usage. Whether PCIs can develop into GCIs depends on the degree of entrenchment of PCIs. Entrenchment essentially refers to the routinization and schematization of associations, which is a kind of cognitive process taking place in individual minds. Schmid (2020, p. 6) points out, “repetition is conducive to entrenchment because it drives routinization and schematization.” The degree of entrenchment is thought to correlate with the frequency of repeated usage (Schmid, 2007). Frequency of repeated usage is an important determinant of entrenchmentFootnote 12 (Bybee, 2006, 2010; Divjak and Caldwell-Harris, 2015; Schmid, 2017, 2020). The higher the frequency of repeated usage is, the higher the degree of entrenchment is (Schmid 2020). “What is cognitively salient, because it has been entrenched very strongly by frequent repetition, is more likely to come to mind in a given usage situation than what is less salient” (Schmid, 2020, p. 223). Highly entrenched PCIs are more likely to become salient, developing into GCIs.
Third, conventionalization is a process of the collective society. Entrenched GCIs are conventionalized into new coded meanings. Conventionalization refers to the agreement shared by all the members of a speech community, involving the usualization and diffusion of utterance types, which is a social process. Frequency of repeated usage is conducive to conventionalization, because “the more frequently an utterance type is repeated, the more it will become usualized and diffused, and the more stable it is likely to be” (Schmid, 2020, p. 121–122). The process of conventionalization is achieved through the usage of repeated activities in the event, affected by the entrenchment process in individual minds (Schmid, 2020). Once entrenched GCIs are widely accepted by all the members of a speech community, they are gradually conventionalized into collective minds and eventually develop into new coded meanings.
Fourth, the entrenchment-conventionalization process of meaning change reflects a process from individual cognition to collective society. Entrenchment is the prerequisite for conventionalization, i.e., only highly entrenched GCIs can be conventionalized. Conventionalization is the inevitable result of entrenchment, because highly entrenched GCIs are more likely to form routine practices used by all the members of a speech community, and eventually become new coded meanings of linguistic expressions. In DCPM, there is a diachronic path from PCIs to GCIs to new coded meanings by way of entrenchment and conventionalization. Based on DCPM, we conduct a case study of the noun phrase hóngyǎn (红眼)Footnote 13 to investigate the diachronic change of conversational implicatures.
Research methodology
In this study, hóngyǎn refers to “eyes whose color is red” as the coded meaning. Based on the CCLFootnote 14 (Center for Chinese Linguistics at Peking University; http://ccl.pku.edu.cn:8080/ccl_corpus/) corpus (Zhan et al., 2003, 2019), this study constructs the hóngyǎn corpus to investigate diachronic change. The method of constructing a small-scale corpus for specific purposes by extracting data from a large diachronic corpus follows previous studies (see Biber, 1993; Leech et al., 2009; McEnery and Hardie, 2012; Petré, 2016; Xu, 2019). For example, Petré (2016) employs a self-built corpus derived from Early English Books Online and Eighteenth Century Collections Online to examine the grammaticalization of [BE going to INF].
The construction of the hóngyǎn corpus based on the CCL corpus aims to address the need for both fine-grained periodization of the diachronic corpus and the precise tokens of each sub-corpus. The hóngyǎn corpus in this study includes 2,983,835 tokens, which is divided into three sub-corpora: the ancient Chinese corpus (97,090 tokens), the modern Chinese corpus (39,212 tokens), and the contemporary Chinese corpus (2,847,533 tokens) (see Table 2).
Therefore, the research data of hóngyǎn is collected from ancient Chinese, modern Chinese, and contemporary Chinese. There are 2460 tokens of hóngyǎn collected in total, including 64 tokens in ancient Chinese, 59 tokens in modern Chinese, and 2337 tokens in contemporary Chinese. After excluding non-hóngyǎn noun phrasesFootnote 15and repeated data, there are 1885 effective tokens of hóngyǎn. Among these 1885 tokens, 286 tokens express the coded meaning “eyes whose color is red” and 1599 tokens express PCIs (see Table 3).
This study adopts the corpus method and the combination of quantitative and qualitative analysis to reveal that PCIs develop into GCIs and eventually become new coded meanings from the perspective of DCPM. The quantitative analysis primarily uses descriptive statistics, including adjusted frequency and percentage measures, to present the distribution of hóngyǎn’s meanings. The qualitative analysis mainly covers ad-hoc inferences, the entrenchment process, and the conventionalization process of hóngyǎn’s meanings. We focus on the diachronic change of these 1599 tokens of hóngyǎn. For example,
(5) 冤家见面, 分外红眼。Footnote 16
Yuānjiā jiànmiàn, fènwài hóngyǎn.
When the enemies met, they became infuriated.
(From Kang Xi Xia Yi Zhuan《康熙侠义传》)
According to DCPM, hóngyǎn in (5) expresses the PCI “become infuriated” via ad-hoc inferences in the specific contextFootnote 17. It involves the act of someone being infuriated, undergoing the process of metonymization. In the self-built corpus, hóngyǎn conveys 15 PCIs in total, namely “pinkeye,” “jealous,” “overnight,” “become infuriated,” “red from crying,” “red-eye effect,” “fatigue,” “game profession,” “grandchild,” “infrared guidance device style,” “drunkenness,” “greed,” “temptation,” “wine,” and “beautiful women.”
It is generally acknowledged that the most frequently used statistics in corpus linguistics are frequencies of (co-)occurrence of linguistic variables. However, frequencies of (co-)occurrence in isolation may be not sufficient (Egbert and Burch, 2023; Egbert and Burch and Biber, 2020; Gries, 2008, 2020). Therefore, this study uses the adjusted frequencyFootnote 18 measure as the main analytical method, because it takes into consideration the degree of dispersion of relevant linguistic variables, i.e., how evenly and widely a word or phrase is distributed across sub-corpora of a corpus (Brezina, 2018; Egbert and Burch, 2023; Gries, 2008, 2020). Using the adjusted frequency measure makes the result more rigorous and more convincing. In this study, the approach of calculating the adjusted frequency measure is to multiply the normalized frequency by the Deviation of Proportions (DP) value. Adjusted frequency can be expressed as:
In this formula, adjusted frequency can be explained in two aspects. First, normalized frequency can be measured by dividing the tokens of a word or phrase in the corpus by the size of the corpus, and then multiplying the result by the basis for normalization. Normalized frequency is expressed as:
For example, the normalized frequency of the PCI “pinkeye” in the hóngyǎn corpus is calculated as follows:
Second, DP is a measure proposed by Gries (2008) which compares “the expected distribution of a word or phrase in different corpus parts with the actual distribution” (Brezina, 2018, p. 52). DP is expressed as:
The observed proportions are to divide the tokens of a word or phrase in the corpus parts by its total tokens in the whole corpus. The expected proportions are to divide the size of the corpus parts by the total tokens in the whole corpus. The number 2 represents a fixed value. By taking the absolute differences between the observed and the expected proportions, summing them up, and dividing by 2, we obtain the DP measure (Brezina, 2018; Gries, 2008). For instance, calculating DP of the PCI “pinkeye” in the hóngyǎn corpus is shown in Table 4.
In Table 4, DP of the PCI “pinkeye” in the hóngyǎn corpus is calculated as follows:
Therefore, we can calculate the adjusted frequency of repeated usage of the PCI “pinkeye” in the hóngyǎn corpus:
It can be seen that the PCI “pinkeye” occurs with an adjusted frequency of 6.827 per million characters. Figure 2 shows 15 PCIs of hóngyǎn with the adjusted frequency of repeated usage.
Fifteen PCIs of hóngyǎn with the adjusted frequency of repeated usage.
In Fig. 2, the adjusted frequency of repeated usage of 15 PCIs of hóngyǎn in the self-built corpus from high to low is: “pinkeye,” “jealous,” “overnight,” “become infuriated,” “red from crying,” “red-eye effect,” “fatigue,” “game profession,” “grandchild,” “infrared guidance device style,” “drunkenness,” “greed,” “wine,” “temptation,” and “beautiful women.” As shown in Fig. 2, “pinkeye,” “jealous,” “overnight,” “become infuriated,” “red from crying,” and “red-eye effect” have relatively high adjusted frequency of repeated usage, and the remaining nine PCIs have relatively low adjusted frequency of repeated usageFootnote 19.
By combining authentic data of hóngyǎn with DCPM, the diachronic process of meaning change can be better represented, so as to contribute to revealing the connection between different types of meanings. In this study, an alternative account for the diachrony of conversational implicatures is provided from a diachronic perspective. The following section will discuss ad-hoc inferences of PCIs of hóngyǎn.
Ad-hoc inferences of PCIs of hóngyǎn
Drawing upon the self-built corpus, this study revisits hóngyǎn (red eyes in English) to see how its meanings are diachronically changed. In the following subsections, we elaborate on ad-hoc inferences of PCIs of hóngyǎn in ancient Chinese, modern Chinese, and contemporary Chinese, respectively.
Ancient Chinese
As mentioned above, the coded meaning is the starting point of diachronic change. In the Ming Dynasty (1368–1644), hóngyǎn appeared for the first time, expressing its coded meaning “eyes whose color is red,” which is exemplified in the following example.
(6) 搭拉两个耳, 一尾扫帚长。青毛生锐气, 红眼放金光。
Dālā liǎnggè ěr, yī wěi sàozhǒu cháng. Qīngmáo shēng ruìqì, hóngyǎn fàng jīnguāng.
With two drooping ears and a tail like a long broom, green fur radiates sharpness and red eyes gleam with golden light.
(From Xi You Ji 《西游记》)
Except for the coded meaning “eyes whose color is red,” hóngyǎn expresses the PCIs in the Qing Dynasty (1636–1912) such as “pinkeye” in (7) and “become infuriated” in (8).
(7) 十三四岁出天花, 才得了个烂红眼, 罗圈腿, 浑身上下大麻子。
Shísānsì suì chū tiānhuā, cái dé liǎo gè làn hóngyǎn, luóquāntuǐ, húnshēn shangxià dàmázǐ.
At the age of thirteen or fourteen, when smallpox appeared, he got pinkeye, bowlegs, and pockmarks all over.
(From San Xia Jian 《三侠剑》)
(8) 张德寿一看少爷, 分外红眼, 说道: “你要什么刀? ”
Zhāng Déshòu yī kàn shàoyé, fènwài hóngyǎn, shuō dao: “Nǐ yào shénme dāo?”
Zhang Deshou looked at the young master, became infuriated and said, “What kind of knife did you want?”
(From San Xia Jian 《三侠剑》)
According to DCPM, based on ad-hoc inferences, hóngyǎn in (7) represents the PCI “pinkeye” through metaphtonymization in the specific context. It metonymically refers to a person’s eyes turning red, and then metaphorically develops into the disease, reflecting the interaction between metonymy and metaphor. In (8), the PCI of hóngyǎn means “become infuriated” via ad-hoc inferences, which is the intra-domain mapping in that hóngyǎn involves the act of someone being infuriated through the process of metonymization. The following examples illustrate that hóngyǎn expresses the PCIs “fatigue” and “jealous” in different contexts, respectively.
(9) 自己坐到四更敲一双红眼皆做梦睡到东方天将晓秀英女子还未醒起来……
Zìjǐ zuò dào sìgèng qiāo yīshuāng hóngyǎn jiē zuòmèng shuì dào dōngfāng tiān jiāngxiǎo xiùyīng nǚzǐ hái wèi xǐng qǐlái……
I sat at the fourth watch with fatigue. I dreamed and fell asleep until the dawn, and Xiuying has not woken up……
(From Xiao Ba Yi 《小八义》)
(10) 不该拿与孩子耍? 只恨拿不到他屋里。头里叫着, 想回头也怎的, 恰似红眼军抢将来的, 不教一个人儿知道。
Bù gāi ná yǔ háizi shuǎ? Zhǐ hèn ná bù dào tā wū lǐ. Tóu lǐ jiào zhe, xiǎng huítóu yě zěn dì, qiàsì hóngyǎn jūn qiǎng jiāng lái de, bù jiào yīgè rén ér zhīdào.
Shouldn’t it be taken to play with the child? I only regret not being able to take it into his room. It’s calling out ahead, and even if I want to turn back, what can I do? It’s just like the jealous army rushing in, not letting a person know.
(From Jin Ping Mei 《金瓶梅》)
In (9), “sìgèng” indicates the time period from 1:00 a.m. to 3:00 a.m. Based on the specific context of “zuò dào sìgèng” and ad-hoc inferences, hóngyǎn denotes the PCI “fatigue.” It metonymically refers to someone being tired from staying up late. In (10), the PCI “jealous” arises from ad-hoc inferences and relates to the act of someone being jealous, which is metonymic. The following Fig. 3 summarizes the distribution of tokens of PCIs of hóngyǎn in ancient Chinese. Hóngyǎn in ancient Chinese generates four PCIs: “become infuriated,” “jealous,” “pinkeye,” and “fatigue” through metaphtonymization and metonymization.
The distribution of tokens of PCIs of hóngyǎn in ancient Chinese.
Modern Chinese
Compared with PCIs of hóngyǎn in ancient Chinese, a new PCI “red from crying” is added in modern Chinese, while “become infuriated,” “jealous,” “pinkeye,” and “fatigue” in ancient Chinese are repeatedly used in modern Chinese.
(11) “你哭完了没有? 我完了! ”王德抹着红眼。
“Nǐ kū wán liǎo méiyǒu? Wǒ wán liǎo!” Wáng Dé mǒzhe hóngyǎn.
“Have you finished crying? I am done!” Wang De wipes his eyes red from crying.
(From Lao Zhang De Zhe Xue 《老张的哲学》)
The PCI of hóngyǎn in (11) is “red from crying” via ad-hoc inferences. It involves the act of a person whose eyes are red from crying, going through the process of metonymization. In modern Chinese, other PCIs, i.e., “become infuriated,” “jealous,” “pinkeye,” and “fatigue,” are further strengthened through repeated usage in different contexts. Based on the analysis of the contexts in which PCIs appear, the distribution of tokens of PCIs of hóngyǎn in modern Chinese is shown in Fig. 4. It is indicated that hóngyǎn in modern Chinese conveys five PCIs: “become infuriated,” “red from crying,” “jealous,” “pinkeye,” and “fatigue” through metonymization and metaphtonymization.
The distribution of tokens of PCIs of hóngyǎn in modern Chinese.
Contemporary Chinese
Compared with PCIs of hóngyǎn in ancient and modern Chinese, ten new PCIs such as “overnight,” “red-eye effect,” “infrared guidance device style,” “greed,” “temptation,” “drunkenness,” “game profession,” “wine,” “grandchild,” and “beautiful women,” are added in contemporary Chinese, while the rest of PCIs, e.g., “become infuriated,” “red from crying,” “jealous,” “pinkeye,” and “fatigue,” are further strengthened by way of repeated usage in contemporary Chinese. The following examples demonstrate that hóngyǎn conveys the PCIs “overnight,” “red-eye effect,” and “infrared guidance device style” in different contexts, respectively.
(12) 据悉, “红眼”航班在国际上很普遍, 并受到广泛欢迎。
Jùxī, “hóngyǎn” hángbān zài guójì shang hěn pǔbiàn, bìng shòudào guǎngfàn huānyíng.
It is reported that an “overnight” airplane flight is very common internationally and is widely welcomed.
(From Xin Wen Bao Dao 《新闻报道》)
(13) 另两家公司还增加了让用户在网页上修改图片的功能, 如消除红眼、改变大小和显示效果等服务。
Lìng liǎngjiā gōngsī hái zēngjiā le ràng yònghù zài wǎngyè shàng xiūgǎi túpiàn de gōngnéng, rú xiāochú hóngyǎn, gǎibiàn dàxiǎo hé xiǎnshì xiàoguǒ děng fúwù.
The other two companies have also added functions that allow users to modify images on web pages, such as removing red-eye effect, changing the size, and displaying effects.
(From Xin Wen Bao Dao 《新闻报道》)
(14) 1983年起, CIA正式开始支援游击队, 提供FIM-43红眼飞弹……
1983 nián qǐ, CIA zhèngshì kāishǐ zhīyuán yóujīduì, tígōng FIM-43 hóngyǎn fēidàn……
Since 1983, the CIA officially began supporting guerrilla forces, providing FIM-43 infrared guidance device style missiles……
(From Zhong Wen Wei Ji 《中文维基》)
In (12), according to Wikipedia, “hóngyǎn” hángbān (red-eye flight in English) refers to an overnight flight, which is generally designed to improve aircraft utilization and use night aircraft availability to schedule flights in order to reduce flight costs. The passengers have red eyes, which can be caused or aggravated by late-night travel. Based on the specific context and encyclopedic knowledge, hóngyǎn in (12) can mean the PCI “overnight” via ad-hoc inferences. This PCI is the inter-domain mapping from the human-body domain to the non-human-body domain, undergoing metaphorization in the specific context. Example (13) illustrates that based on ad-hoc inferences, the hearer infers that the PCI of hóngyǎn is “red-eye effect” in the given context, which is metaphorical. In (14), hóngyǎn metaphorically denotes the PCI “infrared guidance device style” by way of ad-hoc inferences. Similarly, the following three examples show three new PCIs of hóngyǎn in contemporary Chinese.
(15) 今天, 他在外圈巴着红眼看了半天, 直到快晌午清静下来才凑到前边。
Jīntiān, tā zài wàiquān bāzhe hóngyǎn kàn le bàntiān, zhídào kuài shǎngwu qīngjìng xiàlái cái còu dào qiánbian.
Today, he stared at the outer circle with greed for a long time, and didn’t come to the front until it was quiet around noon.
(From Ding Dai Gou Chen《顶戴钩沉》)
(16) 四区人说二月妹子是红眼媚狐, 看样子名不虚传, 肚子里的骚水当真不少。
Sì qū rén shuō èryuè mèizi shì hóngyǎn mèi hú, kàn yàngzi míngbùxūchuán, dùzi lǐ de sāo shuǐ dàngzhēn bù shǎo.
People from four districts said that the girl Eryue was full of temptation. She seemed to live up to her reputation, and there was a lot of water in her stomach. (From Li Shi De Tian Kong 《历史的天空》)
(17) 我也时常经历这样的酒会, 遇着酒后红眼的人, 含着泪倾述自己在书法圈的不易……
Wǒ yě shícháng jīnglì zhèyàng de jiǔhuì, yùzhe jiǔhòu hóngyǎn de rén, hánzhe lèi qīngsù zìjǐ zài shūfǎ quān de bùyì……
I also often experience such drinking parties, meeting people with drunkenness. With tears in their eyes, they pour out the struggles they’ve faced in the calligraphy circle……
(From Wei Xin Gong Zhong Hao 《微信公众号》)
In (15), (16) and (17), the speaker employs hóngyǎn to convey meanings through ad-hoc inferences in different contexts, and the hearer infers that the PCIs of hóngyǎn are “greed,” “temptation”Footnote 20 and “drunkenness” respectively. Hóngyǎn in (15) metonymically refers to someone being greedy. In (16), it involves the act of a person being full of temptation, which is metonymic. In (17), it metonymically relates to the act of someone’s eyes turning red after being drunk. Moreover, hóngyǎn also expresses four new PCIs in contemporary Chinese.
(18) 原来都是哪有任务跟哪练, 我新手。原来人说红眼单刷很强……
Yuánlái dōu shì nǎ yǒu rènwù gēn nǎ liàn, wǒ xīnshǒu. Yuánlái rén shuō hóngyǎn dānshuā hěn qiáng……
It used to be that wherever there was a mission, I would follow to train. I’m a newbie. People used to say that Red Eye (game profession) is very strong at solo grinding……
(From She Qu Wen Da 《社区问答》)
(19) 就着这份烤鸡皮, 请再给我来一杯红眼! (红眼: 冰镇鲜啤与鲜榨番茄汁1:1完美融合, 除了让啤酒入口更加甘醇, 番茄中含有的水溶性成分能够迅速代谢酒精, 从而有效减缓酒醉速度, 所以女生们喝酒的时候不要把红眼给忘了哦!)
Jiù zhe zhè fèn kǎo jīpí, qǐng zài gěi wǒ lái yì bēi hóngyǎn! (Hóngyǎn: bīngzhèn xiānpí yǔ xiānzhà fānqié zhī 1:1 wánměi rónghé, chúle ràng píjiǔ rùkǒu gèng jiā gānchún, fānqié zhōng hányǒu de shuǐróngxìng chéngfèn nénggòu xùnsù dàixiè jiǔjīng, cóngér yǒuxiào jiǎnhuǎn jiǔzuì sùdù, suǒyǐ nǚshēngmen hējiǔ de shíhou bùyào bǎ hóngyǎn gěi wàngle ó!)
With this plate of grilled chicken skin, please bring me another Red Eye (wine)! (Red Eye: a perfect blend of chilled fresh beer and freshly squeezed tomato juice in a 1:1 ratio. Not only does it make the beer taste smoother, but the water-soluble compounds in tomatoes can quickly metabolize alcohol, effectively slowing down the rate of intoxication. So, ladies don’t forget to order Red Eye when you drink!)
(From Wei Xin Gong Zhong Hao 《微信公众号》)
(20) 两人找到村长, 觉得特别面熟, 问他在那边有什么直系亲属, 村长说老红眼儿是他没出五服的四姨。
Liǎng rén zhǎodào cūnzhǎng, juéde tèbié miànshú, wèn tā zài nàbiān yǒu shénme zhíxì qīnshǔ, cūnzhǎng shuō lǎo hóngyǎn ér shì tā méi chū wǔfú de sì yí.
The two found the village chief and felt he looked particularly familiar. They asked if he had any direct relatives over there. The village chief said that his old grandchild was his fourth aunt within five degrees of kinship.
(From Zuo Jia Wen Zhai 《作家文摘》)
(21) 自古红眼多搏命。
Zì gǔ hóngyán duō bó mìng.
Beautiful women often have tragic fates.
(From Wei Bo 《微博》)
In (18), based on ad-hoc inferences, the hearer infers hóngyǎn means the PCI “game profession,” going through metaphorization in the special context of the game. In (19), hóngyǎn metaphorically means the PCI “wine” via ad-hoc inferences, specifically a mix of chilled fresh beer and freshly squeezed tomato juice. Examples (20) and (21) illustrate that, drawing on the specific contexts and encyclopedic knowledge, the PCIs of hóngyǎn are respectively “grandchild”Footnote 21 and “beautiful women,” derived through ad-hoc inferences and viewed as metaphorical in certain folklore or cultural traditions. In contemporary Chinese, other PCIs, i.e., “become infuriated,” “red from crying,” “jealous,” “pinkeye,” and “fatigue,” are further strengthened via repeated usage in specific contexts. The following Fig. 5 displays the distribution of tokens of PCIs of hóngyǎn in contemporary Chinese. Hóngyǎn in contemporary Chinese conveys 15 PCIs through metonymization, metaphorization, and metaphtonymization.
The distribution of tokens of PCIs of hóngyǎn in contemporary Chinese.
To sum up, by investigating the diachronic distribution of tokens of PCIs of hóngyǎn in ancient, modern and contemporary Chinese, the conclusion can be drawn that hóngyǎn conveys a total of 15 PCIs: “pinkeye,” “overnight,” “jealous,” “become infuriated,” “red-eye effect,” “game profession,” “grandchild,” “infrared guidance device style,” “fatigue,” “red from crying,” “drunkenness,” “greed,” “temptation,” “wine,” and “beautiful women” in specific contexts. In the process of ad-hoc inferences of PCIs, metonymization, metaphorization, and metaphtonymization are three main cognitive motivations of these 15 PCIs. Figure 6 presents the adjusted frequency of metonymization, metaphorization, and metaphtonymization in the process of ad-hoc inferences of PCIs of hóngyǎn. In Fig. 6, three cognitive motivations have different effects on ad-hoc inferences of PCIs of hóngyǎn. Metaphtonymization accounts for the highest adjusted frequency (adjusted frequencymetaphtonymization = 6.827), metaphorization comes next (adjusted frequencymetaphorization = 4.774), and metonymization comes last (adjusted frequencymetonymization = 3.681). It is clear in this figure that metaphtonymization plays a dominant role in the cognitive motivations of PCIs of hóngyǎn.
Adjusted frequency of metonymization, metaphorization, and metaphtonymization in the process of ad-hoc inferences of PCIs of hóngyǎn.
The entrenchment process of PCIs of hóngyǎn
According to DCPM, entrenchment is a process of individual cognition. The degree of entrenchment of PCIs is determined by the adjusted frequency of repeated usage. If the adjusted frequency of repeated usage of PCIs is relatively high, PCIs are easy to be entrenched into GCIs. On the contrary, if the adjusted frequency of repeated usage is relatively low, PCIs are difficult to be entrenched into GCIs. The diachronic distribution of tokens of every PCI of hóngyǎn is displayed, as indicated in Table 5.
From a diachronic perspective, it can be seen from Table 5 that the total tokens of PCIs of hóngyǎn in ancient, modern, and contemporary Chinese are 7, 13, and 1579, accounting for 0.44%, 0.81%, and 98.75%, respectively. Therefore, the change process of PCIs of hóngyǎn is mainly manifested in contemporary Chinese. The following Fig. 7 demonstrates the adjusted frequency of repeated usage in the entrenchment process.
Adjusted frequency of repeated usage in the entrenchment process.
According to DCPM, the degree of entrenchment is determined by the adjusted frequency of repeated usage of PCIs. As shown in Fig. 7, “pinkeye,” “jealous,” “overnight,” “become infuriated,” “red from crying,” and “red-eye effect” have a high degree of entrenchment due to their relatively high adjusted frequency of repeated usage, with a total adjusted frequency of repeated usage of approximately 93%. On the contrary, “fatigue,” “game profession,” “grandchild,” “infrared guidance device style,” “drunkenness,” “greed,” “wine,” “temptation,” and “beautiful women” have a low degree of entrenchment, because their adjusted frequency of repeated usage is relatively low, accounting for a total adjusted frequency of repeated usage of about 7%.
This study first discusses six PCIs: “pinkeye,” “jealous,” “overnight,” “become infuriated,” “red from crying,” and “red-eye effect” with a high degree of entrenchment. The diachronic distribution of PCIs of hóngyǎn with a high degree of entrenchment is depicted in Fig. 8.
The diachronic distribution of PCIs with a high degree of entrenchment.
It is indicated in Fig. 8 that the diachronic distribution of these six PCIs of hóngyǎn with a high degree of entrenchment is different. “Pinkeye,” “jealous,” and “become infuriated” have undergone a complete entrenchment process in ancient-modern-contemporary Chinese. “Red from crying” first appears in modern Chinese, being entrenched in modern and contemporary Chinese. “Overnight” and “red-eye effect” do not appear in both ancient and modern Chinese, thus lacking the entrenchment process in both ancient and modern Chinese. However, “overnight” is frequently used in contemporary Chinese with a peak value, and it may be associated with news texts and web texts, influenced by the translation of “red-eye flight,” which lays a foundation for its being entrenched into the GCI.
Different from the above six PCIs, these meanings “fatigue,” “game profession,” “grandchild,” “infrared guidance device style,” “drunkenness,” “greed,” “wine,” “temptation,” and “beautiful women” are difficult to be entrenched into GCIs, because they have relatively low adjusted frequency of repeated usage. For instance, the PCI “temptation” in (16) only appears once, and its degree of entrenchment is extremely low. Therefore, “temptation” is difficult to be entrenched into the GCI. This example tells us that not all PCIs can be entrenched into GCIs, which depends on the adjusted frequency of repeated usage of PCIs.
Based on the above analysis, “pinkeye,” “jealous,” “overnight,” “become infuriated,” “red from crying,” and “red-eye effect” have relatively high adjusted frequency of repeated usage leading to high routinization and schematization, and in this regard these meanings become more salient, being entrenched into GCIs of hóngyǎn. Because “fatigue,” “game profession,” “grandchild,” “infrared guidance device style,” “drunkenness,” “greed,” “wine,” “temptation,” and “beautiful women” have relatively low adjusted frequency of repeated usage, they can hardly develop into GCIs.
The conventionalization process of GCIs of hóngyǎn
Conventionalization has always been a controversial topic in post-Gricean pragmatics. For example, Levinson insists on conventionalization, while Relevance Theorists reject conventionalization. According to DCPM, conventionalization is a process of collective society, achieved through repeated usage activities in usage events and influenced by the entrenchment process in individual minds. It is the process of semanticizing entrenched GCIs.
As stated above, these six PCIs “pinkeye,” “jealous,” “overnight,” “become infuriated,” “red from crying,” and “red-eye effect” have relatively high adjusted frequency of repeated usage, which makes them more routinized. These meanings become more salient, which are easier to be entrenched into GCIs. The more frequently these highly entrenched GCIs are repeated, the more they become usualized and diffused in a speech community, and the more stable they are likely to be. These GCIs are widely accepted by all the members of a speech community, and they are gradually conventionalized into collective minds through repeated usage, eventually developing into new coded meanings.
Nevertheless, the remaining nine PCIs such as “fatigue,” “game profession,” “grandchild,” “infrared guidance device style,” “drunkenness,” “greed,” “wine,” “temptation,” and “beautiful women” have relatively low adjusted frequency of repeated usage, which are difficult to be entrenched into GCIs. According to DCPM, the entrenchment process in individual minds is the prerequisite for the conventionalization process in a speech community. In other words, only highly entrenched GCIs can be conventionalized in a collective society. These nine meanings with a low degree of entrenchment are hard to become usualized and diffused in a speech community. As a result, these meanings can hardly go through the process of conventionalizing/semanticizing meanings.
Based on the aforementioned analysis, six meanings such as “pinkeye,” “jealous,” “overnight,” “become infuriated,” “red from crying,” and “red-eye effect” ultimately develop into new coded meanings of hóngyǎn through the entrenchment-conventionalization process, theoretically qualifying for dictionary inclusion. However, the Modern Chinese Dictionary (7th Edition) only covers three coded meanings of hóngyǎn: “pinkeye,” “jealous,” and “become infuriated.” These three meanings have undergone the entrenchment-conventionalization process in ancient-modern-contemporary Chinese. By contrast, although “overnight,” “red from crying,” and “red-eye effect” exhibit a high degree of entrenchment and conventionalization, they are not included in the Modern Chinese Dictionary. This exclusion may arise from two primary reasons. Firstly, from a diachronic perspective, “overnight” and “red-eye effect” lack the entrenchment-conventionalization process in both ancient and modern Chinese, while “red from crying” lacks this process in ancient Chinese; secondly, the absence of “overnight,” “red from crying” and “red-eye effect” in the dictionary may be attributed to the inherent time lag in dictionary compilation. As a tool for documenting language norms, dictionaries often require prolonged observation and verification to incorporate emerging and stably used meanings into their compilation system. This discrepancy in lexical inclusion underscores the dynamic tension between actual language use and dictionary representation.
This study examines the diachrony of conversational implicatures from the perspective of DCPM. Three viewpoints should be emphasized in this study. First, different types of meanings, such as GCIs and PCIs, are closely related to each other, and there exists a diachronic change of different types of meanings. Post-Gricean pragmatics follows Grice’s view of meaning distinction. By analyzing different contextual meanings of hóngyǎn, a conclusion can be drawn that some GCIs of hóngyǎn develop from its PCIs, which suggests that there is a diachronic change from PCIs to GCIs.
Second, the diachronic change from PCIs to new coded meanings through GCIs is an entrenchment-conventionalization process. Currently, post-Gricean pragmatics adheres to Grice’s principle- or maxim-based theory to infer conversational implicatures. Nevertheless, from a diachronic perspective, PCIs, which are triggered by cognitive motivations (i.e., metonymization, metaphorization, and metaphtonymization) in the process of ad-hoc inferences, develop into GCIs by way of entrenchment, and then entrenched GCIs become new coded meanings by way of conventionalization. In other words, pragmatic meanings can develop into new coded meanings through the entrenchment-conventionalization process rather than pragmatic principles. This study suggests that not all conversational implicatures are inferred based on pragmatic principles.
Third, conventions and contexts coexist with each other. In post-Gricean pragmatics, conventions and contexts are opposed to each other. Based on DCPM, it is argued that meanings develop from highly contextualized to highly conventionalized. And context-dependent PCIs are conventionalized into new coded meanings by way of entrenchment. In other words, the role of convention cannot be denied in the process of meaning change of pragmatic meanings. Contexts are the initial state of meaning change, and conventionalization is a diachronic process that makes contextual meanings become new coded meanings. In the diachronic change of conversational implicatures, DCPM rationally accounts for the relationship between conventions and contexts. Hence, to some extent, the diachronic study of meanings is a supplement to the synchronic study.
Conclusion
This study examines the diachronic development of hóngyǎn, aiming to reveal the diachronic change of conversational implicatures, i.e., PCIs develop into GCIs and ultimately become new coded meanings from a diachronic cognitive pragmatic approach. DCPM reflects that meaning change is the entrenchment-conventionalization process from individual cognition to collective society, and further shows that highly contextualized pragmatic meanings can eventually develop into highly conventionalized coded meanings from a diachronic perspective, and in this regard, conventions and contexts are not exclusive from each other.
It is worth noting that although we exhaustively investigate the development of hóngyǎn from a diachronic perspective, there is a relatively small sample size for hóngyǎn, which is a limitation of this study. Future research should take into account a large number of samples to further validate the diachronic change of conversational implicatures.
Data availability
The datasets analyzed during the current study are available from the corresponding author on reasonable request.
Notes
According to Jaszczolt (2006), post-Gricean pragmatics means all the developments of Grice’s intention-based approach to communication, including neo-Griceans, Relevance Theorists, Bach, and so on.
Q-, I-, and M- principles can be depicted as follows (Huang, 2017b, p. 80–82):
Q-principle:
Speaker: Do not say less than is required (bearing the I-principle in mind).
Addressee: What is not said is not the case.
I-principle:
Speaker: Do not say more than is required (bearing the Q-principle in mind).
Addressee: What is generally said is stereotypically and specifically exemplified.
M-principle:
Speaker: Do not use a marked expression without reason.
Addressee: What is said in a marked way conveys a marked message.
Q-principle is abbreviated from Quantity-principle, involving Grice’s first maxim of Quantity; I-principle is short for Informativeness-principle, covering Grice’s second maxim of Quantity; M-principle is an abbreviation of Manner-principle, including Grice’s maxim of Manner (Levinson, 2000; Huang, 2017b).
A logical form is “a well-formed formula, a structured set of constituents, which undergoes formal logical operations determined by its structure” (Sperber and Wilson, 1995, p. 72).
Sperber and Wilson (1995, p. 158) define the principle of Relevance as:
Every act of ostensive communication communicates a presumption of its own optimal relevance.
Take-for-granted and Not-worth-considering principles can be described as follows (Bach, 1984, p. 44–47):
Take-for-granted principle:
It’s appearing to one that p justifies directly inferring that p provided that (a) it does not occur to one that the situation might be out of the ordinary, and (b) if the situation were out of the ordinary, it probably would occur to one that the situation might be out of the ordinary.
Not-worth-considering principle:
If it occurs to one to do A, one is justified in directly deciding to do A, provided that (a) there occurs to one no thought of a reason to the contrary or of an alternative to A, and (b) such a thought probably would occur if it should.
Although Levinson (2000) is aware of conventions of language use, he still emphasizes that Q-, I-, and M- principles play a vital role in the inference of conversational implicatures.
There are two main research trends in historical pragmatics, namely, pragmaphilology and diachronic pragmatics (Huang, 2017b; Jacobs and Jucker, 1995; Taavitsainen and Jucker, 2010). Pragmaphilology represents a macro approach to focus on the wider changing social and cognitive contexts of the texts in which pragmatic change occurs. Diachronic pragmatics represents a micro approach to concentrate on the interface between a linguistic structure and its communicative use across different historical stages of the same language. Diachronic pragmatics is adopted in this study.
This indicates that there exists a diachronic change from conversational implicatures to conventional implicatures, and therefore, the generation of implicatures is a process of conventionalization (Levinson, 2000). However, Levinson does not systematically explain the change of implicatures from a diachronic perspective, because he states that the diachronic research of implicatures is a supplement to synchronic research (Traugott, 2004).
Invited inferencing means that the speakers evoke implicatures and invite the hearers to infer them, which arises in the context of the flow of speech and interaction and can be deemed to be metonymic. Traugott and Dasher (2002, p. 9) point out, “though speakers and hearers draw from paradigmatically organized sets of constructions, lexical items, and other resources, online production and processing make use of essentially syntagmatic relations and associations. Therefore, invited inferences and metonymic relationships predominate.” “Metonymy is in our view intimately bound up with the mechanism of invited inferencing, and changes identified in the literature as involving metonymic change are typically the outcomes of invited inferencing” (Traugott and Dasher, 2002, p. 282).
“Utterance types are conventionalized links between communicative goals and utterance forms and vice versa” (Schmid, 2020, p. 19).
“Usualization establishes—in the case of innovations—and continually sustains and adapts conventionalized utterance types as onomasiological, semasiological, syntagmatic, cotextual, and contextual regularities of behavior among the members of a community” (Schmid, 2020, p. 92–93).
“Diffusion affects the spread of conventions across the members of a speech community or parts of it, across genres and text types, and across activity types and types of situations” (Schmid, 2020, p. 94).
Usualization and diffusion are not independently of each other, but instead co-operate (Schmid, 2020).
Ad-hoc inferences, proposed by Ehmer and Rosemeyer (2018), are important to the study of meaning change. Sweetser (1990) conceptualizes meaning change as metaphorical mapping. Traugott and Dasher (2002) mainly regard meaning change as metonymic mapping, but they do not deny the role of metaphor in meaning change. Professor Traugott, in personal communication, agrees that metonymization and metaphorization are main mechanisms of semantic change, both of which are not mutually exclusive, but rather mutually enhancing. Therefore, DCPM emphasizes that both metonymization and metaphorization play an important part in the process of meaning change.
When we measure frequencies as indicators of different degrees of entrenchment, Schmid (2020, p. 218) points out, “we can choose from so many different ways of counting: not only absolute vs relative frequencies of various kinds and type frequencies vs token frequencies, but also frequencies of elements and patterns on different levels of abstraction and granularity.” In our study, we adopt adjusted frequency. As regards adjusted frequency, please see “Research methodology” for more information.
Relevance Theorists (Kolaiti and Wilson, 2014; Wilson and Kolaiti, 2017) take the noun phrase red eyes as an example and analyze its enriched meanings from a synchronic perspective, yet ignoring the diachronic change of red eyes. On this basis, we conduct a corpus-based diachronic study of hóngyǎn (红眼) to explore its diachronic change, aiming to reveal the diachrony of conversational implicatures.
The CCL corpus is a large-scale diachronic corpus to provide services for Chinese ontology research, which covers diverse genres such as historical texts, biographies, dramas, fiction, spoken discourse, newspapers, films, crosstalk, and online resources. It subsumes two sub-corpora: the ancient Chinese corpus and the modern Chinese corpus. According to the latest data updated in 2024, it includes 5,841,676,206 tokens in both the ancient Chinese corpus (1,094,768,777 tokens) and the modern Chinese corpus (4,746,907,429 tokens). The ancient Chinese corpus is classified according to dynasties, where the corpus is located. For some corpora that cannot be easily classified according to dynasties, the CCL corpus classifies them into other miscellaneous categories. And the modern Chinese corpus is further divided into two parts: modern Chinese and contemporary Chinese. The former deals with the corpus before 1949, and the latter deals with the corpus after 1949. Because of the periodization of the CCL corpus, data for the Zhou Dynasty to the Qing Dynasty is searched in ancient Chinese; data for 1870s–1949 is searched in modern Chinese; data for 1949–present is searched in contemporary Chinese.
Non-hóngyǎn noun phrases mean that hóng and yǎn cannot be regarded as a noun phrase. For example, 孙小红眼波流动 (Sūn Xiǎohóng yǎnbō liúdòng) means that Sun Xiaohong’s eyes are flowing. In this example, hóng and yǎn are not related to each other.
To better exemplify the arguments made in this research, the research data in the process of analysis is exhibited in three ways: Chinese characters, Hànyǔ pīnyīn transcription and free translation (i.e., English translation), so as to better understand the meaning conveyed by linguistic expressions.
Wilson and Kolaiti (2017, p. 155) point out, “we restricted the discourse context to a default of 6 lines before and after the search term. If the default context did not provide enough clues, we expanded the search to a further 10 lines before and after the search term and, if the context was still insufficient, we marked the case as open/inconclusive.”
In personal communication, Professor Egbert suggests that we need to normalize the token count and multiply the result by the Deviation of Proportions (DP) value to get the adjusted frequency in our study.
Given the overall low frequency of tokens of a word or phrase in the corpus, using the adjusted frequency measure makes the study more reliable due to the consideration of dispersion. In corpus linguistics, it is a relative problem for adjusted frequency to decide what counts as high or low to meet the research purpose. As indicated in Fig. 2, we suggest that 40% of the 15 PCIs with adjusted frequency of repeated usage in descending order can be relatively high, and the rest can be relatively low, which can better reveal the relationship between different types of meanings and further achieve the purpose of this study.
In (16), based on the specific context and encyclopedic knowledge, hóngyǎn is collocated with “mèi hú,” activating its denotative meaning “temptation.”
In Chinese folklore, hóngyǎn is often used to refer to the grandchild. This term reflects the distinction between direct and collateral relatives in traditional family concepts.
References
Andersen G (2015) Relevance. In: Aijmer K, Rühlemann C (eds) Corpus pragmatics: a handbook. Cambridge University Press, Cambridge, p 143–168
Bach K (1984) Default reasoning: jumping to conclusions and knowing when to think twice. Pac Philos Quart 65(1):37–58. https://doi.org/10.1111/j.1468-0114.1984.tb00212.x
Bach K (1994) Conversational impliciture. Mind Lang 9(1):124–162. https://doi.org/10.1111/j.1468-0017.1994.tb00220.x
Bach K (1998) Standardization revisited. In: Kasher A (ed) Pragmatics: critical concepts. Routledge, London, p 712–722
Biber D (1993) Representativeness in corpus design. Lit Linguist Comput 8(4):243–257. https://doi.org/10.1093/llc/8.4.243
Brezina V (2018) Statistics in corpus linguistics: a practical guide. Cambridge University Press, Cambridge
Bybee JL (2006) From usage to grammar: the mind’s response to repetition. Language 82(4):711–733. https://doi.org/10.1353/lan.2006.0186
Bybee JL (2010) Language, usage and cognition. Cambridge University Press, Cambridge
Carston R (2002) Thoughts and utterances: the pragmatics of explicit communication. Blackwell, Oxford
Carston R (2009) The explicit/implicit distinction in pragmatics and the limits of explicit communication. Int Rev Pragmat (1):35–62. https://doi.org/10.1163/187731009X455839
Carston R (2019) Ad hoc concepts, polysemy and the lexicon. In: Scott K, Clark B, Carston R (eds) Relevance, pragmatics and interpretation. Cambridge University Press, Cambridge, p 150–162
Carston R (2021) Polysemy: pragmatics and sense conventions. Mind Lang 36(1):108–133. https://doi.org/10.1111/mila.12329
Deo A (2015) Diachronic semantics. Annu Rev Linguist 1(1):179–197. https://doi.org/10.1146/annurev-linguist-030514-125100
Divjak D, Caldwell-Harris CL (2015) Frequency and entrenchment. In: Dabrowska E, Divjak D (eds) Handbook of cognitive linguistics. Mouton de Gruyter, Berlin and Boston, MA, p 53–75
Egbert J, Burch B (2023) Which words matter most? Operationalizing lexical prevalence for rank-ordered word lists. Appl Linguist 44(1):103–126. https://doi.org/10.1093/applin/amac030
Egbert J, Burch B, Biber D (2020) Lexical dispersion and corpus design. Int J Corpus Linguist 25(1):89–115. https://doi.org/10.1075/ijcl.18010.egb
Ehmer O, Rosemeyer M (2018) Inferences in interaction and language change. Open Linguist 4:536–551. https://doi.org/10.1515/opli-2018-0026
Geeraerts D (2010) Theories of lexical semantics. Mouton de Gruyter, Berlin
Goossens L (2003) Metaphtonymy: the interaction of metaphor and metonymy in expressions for linguistic action. In: Dirven R, Pörings R (eds) Metaphor and metonymy in comparison and contrast. Mouton de Gruyter, Berlin, p 349–377
Grice P (1989) Studies in the way of words. Harvard University Press, Cambridge, MA
Grice P (2001) Aspects of reason. Oxford University Press, Oxford
Gries ST (2008) Dispersions and adjusted frequencies in corpora. Int J Corpus Linguist 13(4):403–437. https://doi.org/10.1075/ijcl.13.4.02gri
Gries ST (2020) Analyzing dispersion. In: Paquot M, Gries ST (eds) A practical handbook of corpus linguistics. Springer, Berlin and New York, p 99–118
Huang Y (2014) Pragmatics. Oxford University Press, Oxford
Huang Y (2017a) Truth-condition-contributing conversational implicatures, intrusive constructions, and neo-Gricean pragmatics. Foreign Lang Teach Res 49(5):643–662
Huang Y (2017b) Implicitness in the lexis: lexical narrowing and neo⁃Gricean pragmatics. In: Cap P, Dynel M (eds) Implicitness: from lexis to discourse. John Benjamins, Amsterdam, p 67–94
Jacobs A, Jucker AH (1995) The historical perspective in pragmatics. In: Jucker AH (ed) Historical pragmatics: pragmatic developments in the history of English. John Benjamins, Amsterdam, p 3–33
Jaszczolt K (2006) Defaults in semantics and pragmatics. https://plato.stanford.edu/entries/defaults-semantics-pragmatics/
Kolaiti P, Wilson D (2014) Corpus analysis and lexical pragmatics: an overview. Int Rev Pragmat 6(2):211–239. https://doi.org/10.1163/18773109-00602002
Leech G, Hundt M, Mair C, Smith N (2009) Change in contemporary English: a grammatical study. Cambridge University Press, Cambridge
Levinson SC (1979) Pragmatics and social deixis: reclaiming the notion of conventional implicature. In: Kingston J, Sweetser E, Collins J, Kawasaki H, Manley-Baser J, Marschak DW, O’Connor C, Shaul D, Tobey M, Thompson H, Turner K (eds) Proceedings of the fifth annual meeting of the Berkeley Linguistics Society. Berkeley Linguistics Society, Berkeley, p 206–223
Levinson SC (2000) Presumptive meanings: the theory of generalized conversational implicature. Cambridge University Press, Cambridge
Levinson SC (2024) The dark matter of pragmatics: known unknowns. Cambridge University Press, Cambridge
McEnery T, Hardie A (2012) Corpus linguistics: method, theory and practice. Cambridge University Press, Cambridge
Petré P (2016) Unidirectionality as a cycle of convention and innovation. Micro-changes in the grammaticalization of [BE going to INF]. Belg J Linguist 30(1):115–146. https://doi.org/10.1075/bjl.30.06pet
Schmid H (2007) Entrenchment, salience, and basic levels. In: Geeraerts D, Cuyckens H (eds) The Oxford handbook of cognitive linguistics. Oxford University Press, Oxford, p 117–138
Schmid H (2014) Lexico-grammatical patterns, pragmatic associations and discourse frequency. In: Herbst T, Schmid H, Faulhaber S (eds) Constructions collocations patterns. Mouton de Gruyter, Berlin and New York, p 239–293
Schmid H (2015) A blueprint of the Entrenchment-and-Conventionalization Model. Yearb Ger Cogn Linguist Assoc 3(1):3–25. https://doi.org/10.1515/gcla-2015-0002
Schmid H (2016) Why cognitive linguistics must embrace the pragmatic and social dimensions of language and how it could do so more seriously. Cogn Linguist 27(4):543–557. https://doi.org/10.1515/cog-2016-0048
Schmid H (2017) A framework for understanding linguistic entrenchment and its psychological foundations. In: Schmid H (ed) Entrenchment and the psychology of language learning: how we reorganize and adapt linguistic knowledge. APA and Walter de Gruyter, Boston, MA and Berlin, p 7–36
Schmid H (2020) The dynamics of the linguistic system: usage, conventionalization and entrenchment. Oxford University Press, Oxford
Schmid H (2023) Multi-dimensional regularity analysis: how the Entrenchment-and-Conventionalization Model can be applied to corpus data. Yearb Ger Cogn Linguist Assoc 11(1):23–56. https://doi.org/10.1515/gcla-2023-0003
Sperber D, Wilson D (1995) Relevance: communication and cognition. Blackwell, Oxford
Sweetser E (1990) From etymology to pragmatics: metaphorical and cultural aspects of semantic structure. Cambridge University Press, Cambridge
Taavitsainen I, Jucker AH (2010) Trends and developments in historical pragmatics. In: Jucker AH, Taavitsainen I (eds) Historical pragmatics. Mouton de Gruyter, Berlin, p 3–32
Traugott EC (2004) Historical pragmatics. In: Horn LR, Ward GL (eds) The handbook of pragmatics. Blackwell, Oxford, p 538–561
Traugott EC (2012) Pragmatics and language change. In: Allan K, Jaszczolt K (eds) The Cambridge handbook of pragmatics. Cambridge University Press, Cambridge, p 549–565
Traugott EC (2018) Rethinking the role of invited inferencing from the perspective of interactional texts. Open Linguist 4:19–34. https://doi.org/10.1515/opli-2018-0002
Traugott EC (2019) Whither historical pragmatics? A cognitively-oriented perspective. J Pragmatics 145:25–30. https://doi.org/10.1016/j.pragma.2019.01.006
Traugott EC, Dasher RB (2002) Regularity in semantic change. Cambridge University Press, Cambridge
Vega Moreno RE (2007) Creativity and convention: the pragmatics of everyday figurative speech. John Benjamins Publishing Company, Amsterdam
Verspoor M, Schmid, H (2024) What counts as usage in dynamic usage-based models. In: McManus K (ed) Usage in second language acquisition: critical reflections and future directions. Routledge, New York, p 87–106
Wilson D, Kolaiti P (2017) Lexical pragmatics and implicit communication. In: Cap P, Dynel M (eds) Implicitness: from lexis to discourse. John Benjamins, Amsterdam, p 147–175
Wilson D, Sperber D (2004) Relevance theory. In: Horn LR, Ward GL (eds) The handbook of pragmatics. Blackwell, Oxford, p 607–632
Xu J (2019) Yuliaoku yu huayu yanjiu [Corpora and discourse studies]. Waiyujiaoxueyuyanjiuchubanshe [Foreign Language Teaching and Research Press], Beijing
Zhan W, Guo R, Chang B, Chen Y, Chen L (2019) Beijing daxue CCL yuliaoku de yanzhi [The building of the CCL corpus: its design and implementation] Yuliaokuyuyanxue [Corpus Linguist] 1:71–86
Zhan W, Guo R, Chen Y (2003) The CCL corpus of Chinese texts: 700 million Chinese characters, the 11th Century B.C. - present. http://ccl.pku.edu.cn:8080/ccl_corpus
Zhang S, Zhang Y (2015) Scalar implicature: a Saussurean system-based approach. Lang Sci 51:43–53. https://doi.org/10.1016/j.langsci.2015.05.003
Acknowledgements
We would like to thank the anonymous reviewers and the Editor for their valuable and helpful comments on several revisions to this manuscript. We are also grateful to Professor Stefan Th. Gries and Professor Jesse Egbert for their constructive suggestions on the adjusted frequency measure of our study. This work is supported by the National Social Science Foundation of China (Grant No. 24BYY086) and the Social Science Foundation of Shandong Province of China (Grant No. 24CYYJ06).
Author information
Authors and Affiliations
Contributions
Ruili Su: conceptualization, data collection, methodology, data analysis and interpretation, writing—original draft, writing—review & editing. Yanfei Zhang: conceptualization, methodology, data analysis and interpretation, writing—review & editing.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Ethical approval
This article does not contain activities involving human participation.
Informed consent
This article does not contain any studies with human participants performed by any of the authors.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Su, R., Zhang, Y. Conversational implicature: a diachronic cognitive pragmatic approach. Humanit Soc Sci Commun 12, 867 (2025). https://doi.org/10.1057/s41599-025-05248-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1057/s41599-025-05248-2
