Huddleston and Pullum (2002) defined modality as the speaker’s opinion on whether a particular subject is necessary or possible. A modal verb is an auxiliary verb that implies probability, necessity, inevitability, or inclination. As their name suggests, modal verbs are important for encoding modality. The attributes of modal verbs can be summarized by the acronym NICE, which stands for negation, inversion, code, and emphatic affirmation (Huddleston, 1976, 333). Negation means that modal verbs can combine with the word not, so that the newly made phrase carries a negative meaning. Inversion means that modal verbs can be put in front of the subject to form an interrogative sentence. Code means that modal verbs can replace predicate verbs in sentences. Emphatic affirmation means that modal verbs can help emphasize a point.

Modal verbs differ from other auxiliary verbs in at least three ways (Palmer, 2014). First, a modal verb has the same form, regardless of whether it follows a singular or plural or a first-person, second-person, or third-person subject. Second, modal verbs do not have an infinitive or gerund form. Third, modal verbs have an exclusive relationship with one another. Thus, one should use only one modal verb at a time.

Broadly speaking, modal verbs include central modal verbs (can, could, may, might, must, shall, should, will, and would), marginal modal auxiliary (dare [to], need [to], ought to, and used to), and semi-modal verbs (have to, [had] better, [have] got to, be supposed to, and be going to; Quirk et al., 1985; Biber et al., 1999). Additionally, Biber et al. (1999) established a relationship between pairs of central modals according to a grammatical value based on the expression of time: can and could, may and might, shall and should, will and would.

Modal verbs are regarded as the main carriers of modal expression in many languages, including Chinese (Peng, 2007). Chinese scholars have given modal verbs various names, such as auxiliary verbs (Liu, 1960; Zheng, 2001) or modal auxiliary verbs (Tsang, 1981). However, these names and their corresponding definitions fail to capture the full range of modality between the positive and negative poles. This full range of modality is reflected in Zhu (1996) definition of qingtaidongci (modal verb) with respect to the semantic and stylistic characteristics of Chinese modality. Therefore, in this paper, modal verb also refers to Chinese auxiliaries expressing modality.

The semantics of modal verbs has been the focus of much corpus-based research over the past decade. There are three major ways to classify modal verbs. The bipartite classification (Lyons, 1977) into epistemic and deontic modality seems to be the earliest one. Palmer (2014) further developed the bipartite model into a tripartite classification that includes deontic, dynamic, and epistemic modality. Biber et al. (1999) and Quirk et al. (1985) distinguish between intrinsic and extrinsic modality.

Among all three modal verb classifications, Palmer’s is the simplest to operationalize since it considers both semantic meaning and pragmatic functions. Thus, this paper will adapt Palmer’s classification of modal verbs into three categories: epistemic, deontic, and dynamic.

Epistemic modal verbs are used to express how certain the speaker is about a statement. For example, “what she stated can/might/should/will/may/must be wrong.” Deontic modal verbs are used to express how much responsibility the subject assigns to the object. For example, “you should get the work done by tomorrow” or “I must submit the paper on time.” Dynamic modal verbs can be used to express the ability or willingness of the subject. For example, “he can take this job” or “I will join this team on the project.”

As each modality has a certain magnitude and pragmatic orientation (Halliday, 1994), the speaker can alter the pragmatic orientation of the discourse based on the value implied in the modality expression. Thus, the affirmative and negative of one modal verb might have different values. Halliday (1994) believes that the modality of positive modal verbs (modal operators) can be roughly divided into three levels based on their modal intensity: low, medium, and high. In terms of pragmatic orientation, the lower the value of modal verbs, the weaker and more euphemistic these modal verbs are.

Based on Halliday (1994) value classification and Huddleston and Pullum (2002) semantic analysis of modal verbs, modal verbs in this paper are divided into four groups based on their values. There will be “can/may/could/might” for Value 1, “would” for Value 2, “shall/should” for Value 3, and “must” for Value 4 (Table 1).

According to their values, Chinese modal verbs can also be divided into several groups. Xu (2018) identified twenty-four Chinese model verbs. The formality of GPCs, however, reduces the number of Chinese modal verbs found in the corpus to nineteen. Table 2 shows the categorization of these nineteen Chinese modal verbs according to Li (2018) and Xu (2018), as well as the values of their English counterparts.

Studies on modal verbs tend to focus on written rather spoken aspects. Using computer learner corpora, Aijmer (2002) compared the range and frequency of some key modal words between native English and advanced English learners’ writing using a 52,000-word corpus. The findings show that L2 writers overused modal verbs, a tendency that may be both developmental and interlingual. Chinese speakers who use English as a foreign language may overuse verbs because Chinese is a verb-heavy language (Pease and Cheung, 2018). While Aijmer’s study reports the frequency of different modal verbs in both corpora, it fails to explain why the frequency varies from one modal verb to another. Studies that investigate modal verbs used in oral output tend to use corpora with limited sizes. Park (2019) examined the distribution of modal verbs in a corpus of about 55,000 spoken words and concluded that English learners from Korea produced fewer modal verbs than native speakers, contradicting Aijmer (2002) findings. Römer (2004) conducted a comparative analysis of modal verbs by using spoken British English corpus data and data from English as a foreign language (EFL) textbooks and found discrepancies between authentic English and English textbooks in terms of the use of modal verbs. The modals will, can, and must were overused in the textbooks, while would, could, should, and might were underused.

Frequently regarded as a challenging grammatical structure in English (Palmer, 2014), modal verbs pose one of the greatest challenges for EFL learners (Saeed, 2009; Bensaid, 2016; Cournane and Pérez-Leroux, 2020). The misuse of modal verbs by EFL learners could result in unintended pragmatic consequences in cross-language communication contexts (Hyland and Milton, 1997). Markkanen and Schroder (1997) stressed the importance for EFL leaners to become familiar with the use of modal verbs to avoid potential misunderstanding. Therefore, observing EFL speakers, including interpreters, is imperative for shedding light on the link between use of modal verbs in EFL and effective cross-cultural communication.

Most corpus-based investigations of modal verbs are found in the translation literature instead of interpreting. Kranich’s (2009) study of the translation of scientific texts from English to German revealed that translators tend to use higher-value epistemic modals by using a relatively large corpus of over 500,000 tokens; however, it only examined popular scientific texts. Zhao and Liang (2013) compared the English–Chinese translation of the epistemic verbs may and might between literary and non-literary texts, thus expanding the scope of study to literary translation. Wu (2019) explored the differences between four Chinese translations of Shakespeare’s Measure for Measure in terms of can, may, must, shall, and will. Pei and Li (2018) investigated the distribution of the semantic meanings and values of more than ten Chinese and English modal verbs in thirty-three Chinese civil–commercial legislative instruments and their English translations.

There are few studies on the interpreting of modal verbs. Warchał and Łyda (2009) were the first to study modal verbs in interpreting and collected eighteen student interpreters’ Polish–English and English–Polish consecutive interpretations and examined their transfer of epistemic modal markers. However, with a “translation failure” of about 21% (Polish–English) to 26% (English–Polish), it seems that this research’s primary focus was on L2 learners’ utilization of modal verbs and thus will not provide any insights into the interpreting of modal verbs.

Chinese scholars (Li and Hu, 2013; Li, 2018; Fu and Chen, 2019) have used the GPC corpus to study the shift of modal verbs identified in English interpretation. Their findings suggest that such a shift could be attributed to the need to follow the conventions in the target language. However, none of these studies explored the modal verb distribution in both the ST (source text) and TT (target text), nor did they address the difference in values between modal verbs within the comparable pair. This fails to help us to understand how modal verbs are interpreted in the GPCs. Most of the research has focused on the utilization of modal verbs of strong value and the function of low-value modal verbs in interpretation has been neglected. As one of the most important and difficult grammatical systems in English (Palmer, 2014), modal verbs indicate proficiency in English (Thomas, 1994; Römer, 2004).

This study addresses this research gap by examining the distribution and value of modal verbs in both the ST and TT for a more accurate description of Chinese–English interpreting in terms of modality. Drawing on a self-built corpus of GPC interpreting, this article attempts to conduct a comprehensive intertextual analysis and seeks possible explanations behind the statistics. To this end, this study attempts to answer the following two questions:

Table 3 shows that the proportion of modal verbs in English interpretation is higher than that in Chinese STs. The English corpora contain more modal verb use than the Chinese corpora. Due to its paratactic nature, Chinese uses covert coherence to illustrate grammatical relationships in loosely structured, short, and simple sentences (Cheung, 2009a). In contrast, the English language emphasizes the logic and completeness of sentence structures by using modifiers. Thus, when interpreting Chinese into English, interpreters may have to make explicit in the English target language what is implicit in the Chinese source language (Cheung, 2007, Cheung, 2009a,b, 2016). This could account for the differences in the number of modal verbs between the Chinese and English corpora.

Table 4 shows that the most frequently used Chinese modal verb is “要,” which has the approximate meaning of “should” in English. The second most frequent is “想,” meaning “will” in English.

Adopting from the typology of modality in Halliday (1994) and the scale of modal orientation in Li (2018), all modal verbs were assigned a numerical value from 1 to 4. Table 4 shows that the two most frequently used modal verbs in the Chinese are Value 3, accounting for 55.7% of all modal verbs. A total of 9% of modal verbs are found to be Value 4, which is the lowest of all four types. According to Perkins (1983), there are only sixteen modal verbs in English, while there are twenty-four in Chinese (Xu, 2018). Chinese speakers have more choices of modal verbs than English speakers, and this fact has an important effect on our corpora. The distribution of Chinese modal verbs is not as centralized as English. The figure may be indicative of the preference for more formal language in press conferences on the part of the Chinese premiers, as well as attempts at avoiding ambiguity and conveying toughness.

A total of 2,658 English modal verbs were identified from this corpus of 128,960 words, of which the “will” and “shall” verbs were manually differentiated into modal verbs and auxiliary verbs. Table 5 shows that the most frequently used modal verb is “will,” appearing 1,320 times among 128,960 tokens and making up nearly half of the modal verbs in the interpretation. The proportion of English modal verbs in the TT with the value of 3 increases, while the proportion of modal verbs with the value of 4 decreases, when compared to Chinese modal verbs in the ST. The modal verbs used most frequently after “will” are “can,” “would,” “must,” and “should.” Table 5 shows the frequency of each modal verb in English.

Table 6 shows the similarities and differences between how often modal verbs are used in English-interpreted Chinese GPCs compared to the native-English ones. Frequency in Table 6 refers to the how often a modal verb is used per thousand words of the entire corpus, while percentage refers to how frequently each modal verb is used per hundred modal verbs. A list of the modal verbs investigated by Li and Hu (2013, 28) appears in the Original English GPC column.

Table 6 shows that the overall frequency of modal verbs in English interpretation is 20.06‰, which is 16% higher than the English-speaking GPCs. However, the difference between the usages of each modal verb is even more noticeable.

On the one hand, in interpretation, low-value modal verbs are used less frequently. Modal verbs such as “can,” “could,” and “might” are rarer in the interpretation than in English-speaking GPCs. As Table 6 shows, “could” shows up only 0.47 times every 1,000 words in interpretations of Chinese GPCs, while it shows up much more regularly in English GPCs. This difference is also true of the usage of “might.” However, “may” is an exception to this rule. Although “may” is low-value, it is more frequently used in Chinese GPCs (0.96 times per 1,000 words) than in English-speaking ones (0.47 times per 1,000 words).

On the other hand, modal verbs of a medium or higher value are used much more frequently in the interpretation of Chinese GPCs than in English-speaking ones. As Table 6 shows, “must,” “shall,” and “should” are used more often in interpretation. The results show that the word “should” appears in English translations of Chinese GPCs nearly twice as often as it appears in English press conferences. In the former case, it occurs 2.05 times for every 1,000 words, while for the latter, it appears 1.1 times for every 1,000 words. The biggest gap in the usage of high-value modal verbs is “must.” The data show that “must” appears in the interpretation 2.06 times for every 1,000 words spoken, five times that of the English-speaking GPCs, suggesting that modality value tends to be intensified when interpreting from Chinese into English.

Table 7 shows the differences in the average of modal values among the corpora.

Table 7 shows that the TT has the highest (2.60) average modal value, followed by ST (2.49), while Original English GPC has the lowest (2.17) among the three. The difference between the TT and ST is 0.11. However, the difference between the TT and the Original English GPC is 0.43, or 19.382%. This difference seems to contradict the claim that the need to comply with the conventions of the target language leads to shifts in modal value when interpreting from Chinese into English (Li and Hu, 2013; Li, 2018; Fu and Chen, 2019). A negligible difference in the average of modal value between TT and Original English GPC may support that claim. However, a nearly 20% difference in average modal values between the TT and Original English GPC implies that there could be other reasons to account for an increase of modal values in TT.

Interpreted modal verbs fall into five categories: deletion, amplification, direct translation, switching between affirmation and negation, and using modal verbs of different values. As Table 8 shows, only 39.69% of the modal verbs in the interpretation are directly derived from the ST without value change, while a similar percentage of modal verbs are additions in the TT that did not have a corresponding modal verb in the ST. Additionally, 20.62% of the modal verbs have their value changed in the process of interpreting, and fewer than 1% of the modal verbs are shifted between affirmative and negation. It is worth mentioning that 141 (7.98%) of the modal verbs used in Chinese are omitted in English.

Amplification refers to the addition of a modal verb in the interpretation that is not found in the source language. As shown in Table 9, the interpretation contains a total of 2,658 modal verbs, of which 1,031 are the results of amplification, making up 38.79% of all the modal verbs. Table 9 also shows that high-value modal verbs such as “should,” “will,” and “must” take up the smallest percentage of amplification. Modal verbs that make up the highest proportion are “might” (78.57%), “shall” (80%), and “could” (55.1%).

Two primary reasons may account for the use of amplification. First, when interpreting from Chinese to English, meanings that are implicit in Chinese need to be made explicit so that the interpretation is comprehensible to English speakers; hence, extra words are added (Cheung, 2009a,b), including modal verbs. The following examples illustrate the amplification of modal verbs in the target language.

There are twenty-four cases of switching modal verbs between affirmation and negation. Of the twenty-four conversions, twenty-one are changed from negative in Chinese to positive in English, probably to follow language conventions in English, as in the following example.

Modal value also correlates with levels of politeness (Leech, 1987). Analyzing the difference, if any, between the ST and TT modal values may reveal the level of certainty between the two languages. Analyzing the changes in the value of each modal verb between the ST and TT may provide insights into the handling of modal verbs in settings characterized by cross-cultural and cross-linguistic communication. As Table 10 shows, 548 modal verbs in the TT experienced value changes in comparison to their Chinese counterparts. Of those, 285 are rendered into higher-value modal verbs in the TT, and 262 into lower-value ones. The value difference was calculated by comparing the value in the interpretation to the value in the original text. For example, if the value in the TT is 3, and the value in the ST is 1, the difference in value is 2. The value deviation of modal verbs from the ST to the TT was within the range of −2 to 2. There are no instances where the differences are either 3 or −3, suggesting that there are no cases in which there is an extreme difference in the modal value of the ST and TT. Table 10 shows that in most of the value changes, the value variation between the ST and TT is only 1.

Omission means no trace of the ST modal verb is found in the TT. Results show that 7.98% of the modal verbs in the ST were omitted. Chinese–English interpreting usually uses omission to eliminate redundant words to be idiomatic in English, as the following example shows.