A grammatico-pragmatic analysis of the because X construction: Private expression within public expression

1. Introduction

The word because in English, which is typically followed by a finite clause or an of phrase, conveys a cause or reason, as exemplified in (1a, b):

An innovative use of because has recently emerged that is in use particularly in online communication and colloquial conversation in which a single word or non-clausal phrase directly follows because, as exemplified in (2):

As single words in various grammatical categories such as nouns, adjectives, and interjections can follow because, the construction is called the because X construction. This article investigates the characteristics of this construction, focusing particularly on the status of the X-element from the perspective of the three-tier model of language use (e.g., Hirose 2015). Specifically, it is claimed in accordance with Kanetani (2017) that the X-element serves as a private expression, namely, a speaker’s expression of thought with no intention of communication, although the whole construction is used publicly.¹ Pragmatic effects in response to the proposed structure are also discussed.

The present article is organized as follows. After outlining the research methodology in Section 2, I observe the functional and syntactic properties of the construction as well as the construction’s characteristic expressive structure in Section 3. Section 4 reviews the typological characteristics of English from the perspective of the three-tier model of language use. Section 5 investigates various ways in which the X-elements are construed as private expressions. Sections 6 and 7 explore the motivations for these private expressions.

2. Methods

The analysis in the present article follows a traditional linguistic methodology. First, the grammar of the target constitution is described in detail from both semantic and syntactic points of view based on the grammaticality or acceptability of linguistic data, which are collected from the literature and online sources including tweets, blogs, and corpora. As the because X construction has only recently emerged and come to be recognized, the availability of descriptive data from research papers is limited and the description of its grammar is not sufficient. Therefore, describing the semantic and syntactic characteristics of the construction based on the collected data is essential for analyzing the construction.

After the grammar is described, the construction is analyzed within a certain theoretical framework, and discussions about theoretical implications follow. Specifically, the construction is analyzed by revising and expanding Kanetani’s (2016, 2017, 2019) account of the because X construction from the perspective of the three-tier model of language use, which claims that the X-element functions as a private expression while the entire construction functions as a public expression. The three-tier model is a grammatico-pragmatic theory developed by Hirose (2013, 2015, 2016) as a natural extension from the notion of private and public expressions (Hirose 2000). While the details of the model will be introduced in Section 5, it should be noted here that the grammatico-pragmatic characteristics of a language can only be defined relatively to other languages. In fact, the three-tier model ascribes differences between languages to different combination patterns of the three tiers of language use, namely, the situation construal, situation report, and interpersonal relationship tiers. Therefore, to highlight the grammatico-pragmatic traits of the English language, it is compared to Japanese and the markedness of the construction within the system of English grammar is emphasized.

However, my earlier analyses left open the question of what motivates the proposed expressive structure of the construction. Therefore, comparing the because X construction with other linguistic phenomena of similar expressive structures (an innovative use of kudasai ‘please’ in online communication (Naya 2017) and soliloquy insertion in conversations (Hasegawa 2010)), the present article discusses the significance of the proposed structure and a general metapragmatic strategy behind these expressions.

3. The because X construction

Kanetani (2015, 2016, 2017, 2019) treats the phenomenon from the perspective of construction grammar, where a construction is generally defined as a conventionalized pairing of form and function (Fillmore et al. 1988; Goldberg 1995; Hoffman and Trousdale 2013, among many others). After briefly reviewing the functional properties of the construction in Section 3.1, I identify the formal properties in Section 3.2 and then describe the form-meaning correspondence in Section 3.3.

4. The three-tier model of language use

In Section 3.4, the two-layered expressive structure was proposed, as shown in (16), in which the element in the X-slot crucially serves as a private expression while the entire construction may function as a public expression. This section considers what it means that a private expression is used in a public expression in terms of the three-tier model of language use (Hirose 2015).

As mentioned in Section 2, the three-tier model is a grammatico-pragmatic theory proposed inter alia by Hirose (2015) as a natural extension from the deconstruction of the speaker into the private self as the subject of thinking and the public self as the subject of communicating (Hirose 2000). According to the three-tier model, language use comprises the three tiers listed in (17a-c), and “languages differ as to how the three tiers are combined, according to whether their basic egocentricity lies in the public self or the private self” (Hirose 2015: 123).

In this model, the English language is characterized as follows:

To highlight the characteristics of English, let us compare them with those of Japanese, which is described as follows:

The crosslinguistic difference in the unification pattern of the three tiers is illustrated in Figures 1 and 2, where the bold faces indicate the tiers in which the unmarked deictic center is located, i.e., the default position where deictic expressions are interpreted.⁹

Figure 1.

English as a public-self-centered language.

Figure 2.

Japanese as a private-self-centered language.

Figure 1 illustrates that the situation report tier, where the unmarked deictic center is located, is unified with the situation construal tier in English. Therefore, English speakers need to construe the situation objectively as they report it to others. Conversely, Figure 2 shows that, in Japanese the unmarked deictic center is located in the situation construal tier, which is independent of the unification of the situation report and interpersonal relationship tiers, allowing Japanese speakers to express the situation as they construe it.

Ide (2006) also neatly describes the typological difference. According to Ide, while Japanese speakers tend to view themselves as participants in the situation described, English speakers tend to take the perspective of an omniscient narrator and overview the entire speech event from the outside. The different perspectives of the two languages are illustrated in Figures 3 and 4.¹⁰

Figure 3. Mary gave me this book [English] (Ide 2006: 223).

Figure 4. Mary gave (me this book) [Japanese] (Ide 2006: 222).¹²

As illustrated in Figure 3, an English speaker sees and describes herself on stage as others see her. A Japanese speaker, illustrated in Figure 4, plays the role of a participant on stage as well as the narrator. The speaker, who is embedded in the situation, has to linguistically encode only what is necessary and leaves other elements (me, this book) unspoken. Thus, to describe the same situation, the Japanese speaker would say as follows, with the parenthetical elements not necessarily being expressed.¹¹

In short, English speakers prefer to take an objective perspective from outside of the situation while Japanese speakers prefer to take a subjective perspective from the inside. This observation is compatible with the three-tier model, because the default deictic center in English is the public self, located in the situation report tier (as shown in Figure 1), while that in Japanese is the private self, located in the situation construal tier (as shown in Figure 2).

Another important typological difference Hirose (2015) puts forward as a consequence of the three-tier model is that the unmarked mode of expression is public expression in English and private expression in Japanese. Consider the following contrast:

The unification of the situation construal and situation report tiers in English means that the utterance today is Saturday has a performative structure (cf. Ross 1970), as in (21):

Thus, the unmarked mode of expression in English is considered public expression. In contrast, the Japanese unmarked sentence in (20) cannot have a similar structure to (21), as shown in (22):

Note that all of the Japanese sentences in (19) and (20) convey one and the same propositional content, today is Saturday. The unmarked sentence in (19), however, functions as a private expression by itself; hence, it is incompatible with the performative clause, as shown in (22). Instead, various expressions sensitive to the interpersonal relationship, such as the unmarked sentence-final particle yo, the (super) polite form of the copula desu or degozaimasu, etc., are employed to make the expression public.

With the typological characteristics of English in mind, let us consider the fact that the X-element represents the speaker’s private expression. According to the three-tier model, an English speaker essentially takes a reporter’s (or an objective) perspective and linguistically encodes as much as is necessary to do so; hence, the unmarked mode of expression in English is public expression. The expressions in the X-slot, on the other hand, exhibit the speaker’s subjective construal in that the speaker does not linguistically encode what is already given in his/her consciousness, which is characteristic to languages like Japanese. This claim is in line with Bergs’s (2018) argument that the because X construction is subjective compared with the content causal because-clause construction in the sense of Traugott and Dasher (2004). Crucially, it is not the whole construction that is subjective but only the part that deviates from the norm of the English language. Presumably, the subjectivity Bergs observes in this construction is related to the subjective nature of the X-element. In the following sections, I examine the privateness and subjectivity of the X-element.

5. X-element as private expression

It was observed in Section 3.2 that nouns, adjectives, interjections (including compressed clause), agreement words, and reduced clauses frequently appear in the X-slot. In Section 3.4, the two-layered expressive structure for the because X construction was proposed, which is reproduced here as (23):

In the following subsections, I examine the “privateness” of the expressions that frequently appear in the X-slot and the “publicness” of those that rarely appear.

6. Privateness and content causal relation

In Section 3, I claimed that the meaning of the because X construction is restricted to Sweetser’s (1990) content causal relation. After pointing out in Section 4 that the X-element exhibits a characteristic of a private-self-centered language like Japanese, I observed in Section 5 that the X-element serves as private expression. In this section, I consider how these facts are intertwined along with Kanetani’s (2017) view of a speech act unit as a small discourse. While Sweetser (1990) proposes the three domains in which because functions, Kanetani (2019) claims that the epistemic and speech act because-clauses should be grouped together and that the distinction of the content causal relation from epistemic/speech-act causal relation is crucial. The division is based on how speech act units are formed. A sentence in the content domain performs one speech act as a whole, whereas two independent speech acts are performed in the epistemic/speech-act domain. Compare the following sentences, where the arrows (↑ and ↓) indicate intonation patterns:

Sentences (39a, b) are interrogative sentences with a content because-clause and an epistemic because-clause, respectively. The rising intonation appears at the end of the sentence in (39a), which indicates that the scope of the question encompasses the whole sentence, thus performing a single speech act: The rising intonation in (39b) appears at the end of the main clause. That is, the question scopes over the main clause, while the because-clause independently performs a statement speech act (see Section 3.1; cf. also Lakoff 1987). Thus, the distinction of the epistemic and speech act because-clauses may be reduced to the kind of speech act (e.g., a statement, question, or imperative) being performed in the main clause (Kanetani 2019).

Given the differences in speech act unit formation between the content because-clause construction and the epistemic/speech-act because-clause construction, I recast each speech-act unit here as a “small discourse” to account for the relation between the content reading and the private nature of the X-element in the because X construction. Namely, the content because-clause construction consisting of one speech act unit can be taken as a discourse, which starts with the situation described in the main clause and ends with the situation described in the because-clause. By contrast, in the epistemic/speech-act because-clause construction, there exist two paratactic discourses that are independent of each other. For instance, the content causal sentence the ground is wet because it has rained depicts (as it were) one scene, whereas the epistemic causal sentence it has rained, because the ground is wet depicts two separate scenes, one about raining and the other about the ground being wet. The obligatory comma intonation between the main clause and the epistemic/speech-act because-clause (Sweetser 1990) symbolically represents the discourse boundary or the scene shift.

Together with the notion of small discourse, let us consider the because X sentence in (2), repeated here as (40):

Since the sentence describes a content causal relation, its discourse structure is also assumed to be the same as that of the content because-clause construction. Specifically, the sentence delivers a discourse on the speaker being unable to go out because of his/her homework/sickness. At the beginning of the discourse, the speaker takes the reporter’s perspective, placing his/herself in a situation as a participant, just as with the speaker in Figure 3. That is, the speaker starts the discourse with the unmarked mode of expression in English, playing the role of a narrator, who observes the situation objectively standing on a par with the hearer. As the discourse progresses, however, the speaker switches his/her perspective to a perspective from the inside, describing the situation as a participant, as if he/she jumped into the situation and fused with his/herself on stage, just like the speaker illustrated in Figure 4.

What then makes the speaker switch perspectives? As seen in Section 4, the because X construction itself has a public function (Schnoebelen 2014; Bohmann 2016). Therefore, by starting with the unmarked mode of expression in English, the speaker indicates the publicness of the expression while avoiding the abrupt occurrence of a private expression. Then comes a private expression which indicates his/her own thought expression. As such, it naturally follows that the epistemic/speech-act causal relations are restricted. As mentioned earlier in this section, an epistemic/speech-act because-clause introduces a new discourse independent of the main clause, even though the because-clause appears after the main clause. Thus, the speaker needs to start the new discourse in the unmarked mode.

Some examples may seem problematic for the small-discursive account. Consider the following examples:

In (41a), the main clause subject, the NSF (National Science Foundation), is a third person and is not identical with the speaker who jumps into the situation toward the end of the discourse. In (41b), on the other hand, the [because X] part appears sentence-initially, so it seems difficult to maintain the idea that the speaker switches his/her perspective as the discourse progresses.

Let us first consider example (41a). The actual tweet is linked to a blog written by the same person, where a detailed explanation is given, as in (42):

The first section underlined in (42) “the NSF has canceled the program, at least for this grant cycle” makes virtually the same statement as the main clause in (41a). The other underlined part states the reason: “Congress passed the Coburn Amendment to the Continuing Appropriations Act of 2013, which limits political science funding to research that ‘promotes national security or the economic interests of the United States’”. However, as is clear from the sentence between the two underlined parts, the reason is the author’s opinion. In short, in (41a), using the word politics, the author represents the NSF’s intention and explains it from his own point of view. Thus, in (41a), the author starts the discourse with the narrator’s perspective and then presents his own private expression while maintaining the narrator’s perspective without being fused with any participant in the situation.

The other case we need to consider is (41b), where the [because X] precedes the main clause. When a because-clause appears in sentence-initial position, it is contextually presupposed. To confirm this, consider the following dialogue:

In (43), speaker B’s response to A’s question is anomalous. The response should assert the reason for the ground being wet; nevertheless, the sentence-initial because-clause indicates that the reason is contextually presupposed. By the same token, sentence-initial [ because X] may be considered to be contextually presupposed. In other words, a sentence-initial [because X] may be used only within a context where X is established as (part of) a topic. This can be confirmed by seeing the actual context of use, as COCA allows us to check the context in which the sentence is used. Sentence (41b) appears during an interview on the performance of a space telescope, where the interviewee talks about how distant galaxies can be resolved into individual stars with the telescope. To this extent, sentence (41b) causes no abruption.

7. Significance of the X-element serving as private expression

Thus far, I have claimed that the because X construction has a two-layered expressive structure without considering its motivations. In this section, I examine the meaning of the two-layered structure from metapragmatic strategy perspectives. Some phenomena with the two-layered expressive structure in Japanese have been reported in the literature. First, Naya (2017) investigates an innovative use of kudasai ‘please’ in social networking services, as in (44), which is distinguished from its canonical use with the te-conjunctive form ore-no tooan-o tensakushite kudasai ‘please correct my answer(s)’.

Naya regards the imperative form tensakushiro ‘correct _IMP’ as bearing a private function, which is turned into a public expression by adding kudasai ‘please’. That is, the sentence is interpreted as expressing dual messages, as in (45): the speaker’s wish as expressed by the private expression and the speaker’s request as expressed by the public expression.

Since the construction is used exclusively in the environment of online communication, Naya argues that its use is motivated by this environment, in which some can fulfill the speaker’s wish but others cannot. Those who can correct the speaker’s answers may understand the sentence as an indirect request, while those who cannot may understand it simply as the speaker’s expression of a wish. Naya claims that the use of sentence (44) reflects a metapragmatic strategy of taking various users into consideration. That is, while avoiding being too polite so that the request may not threaten the positive face of the members of the social networking service community, the speaker also indicates negative politeness by indirectly requesting the correction (cf. Brown and Levinson 1987).

Second, Hasegawa (2010: chapter 5) observes that soliloquy sometimes appears in conversations. Observe the following dialogue between a teacher, indicated by H (Higher social status), and a student, indicated by L (Lower social status):

Because of their different social statuses, L is required to use honorifics. Nevertheless, in her second turn, Hee, sensei demo soonandaa, she uses the plain form. Moreover, the exclamatory interjection hee and the exclamatory sentence-final particle daa indicate that the utterance is understood as soliloquy ( Hasegawa 2010: 160). As soliloquy is inserted in a conversation, this may be a case where a private expression appears within a public environment at a discourse level. According to Hasegawa, L in (46) strategically uses the soliloquy in her last turn. While she needs to indicate deference, the use of honorifics necessarily indicates psychological distancing as well. When she wishes to simultaneously express deference and intimacy, she “may temporarily quit the on-going dialogic discourse and switch to soliloquy” (Hasegawa 2010: 162). Hence, Hasegawa views this strategy as a “metapragmatic shift” by which to mitigate the psychological distancing.

In both cases, the use of a private expression within a public forum may be related to the indication of intimacy or solidarity. Although they are examples in Japanese, the same is true in English. Hirose (2013) observes that sentence (47b) conveys a greater sense of closeness than (47a).

According to Hirose (2013: 24), by omitting the subject I, the speaker of sentence (47b) “is describing the situation not from the perspective of an outside reporter, but from that of an inside participant, which he imposes on the addressee; this results in the speaker bringing the addressee closer to him”. Thus, Hirose points out that the subjectless sentence is not compatible with a respectful form of address, as shown in (47b). This might seem contradictory to the fact that the soliloquy in (46) indicates the speaker’s intimacy to the addressee of a higher social status. However, the soliloquy in (46) serves as a private expression, which is not intended to be communicated to but to be overheard by the hearer. In contrast, because of the address terms, (47b) must be considered a public expression addressed to sir or Professor Brown. Therefore, they are essentially distinguished from each other. Hirose (2013) observes that (47b) is a subjective utterance compared with (47a), just like sentences typical of diary English, as in (35). Crucially, without the address terms, the subjectless utterance hope you like it marks friendliness, and to this extent, the use of soliloquy to be overheard in a conversation and the subjectification by omitting the first-person subject both contribute to shortening the distance between the speaker and hearer.

Generally, by using subjective expressions, the speaker brings the hearer closer to him-/herself (Hirose 2013). Private expressions are considered to be highly subjective because they may be used regardless of others (cf. Hirose 2016). In other words, the speaker as a private self reveals his or her “bare” thoughts. The preface to Hasegawa’s (2010) monograph describes the act as follows: “revealing one’s thoughts without interpersonal linguistic devices is a manifestation of trust”. By being attracted to the speaker who reports a situation from the inside, the hearer also has to see the situation from the same perspective as the speaker—namely, the perspective from the inside. Accordingly, a sense of the involvement in the situation may be virtually shared with the hearer, yielding an effect of joint attention. Note that joint attention effects are observed not only in the context of early-stage language acquisition (e.g., Tomasello 1988) but also in certain linguistic uses among adult speakers (e.g., Cheshire 1996). It is also worthwhile pointing out a similarity between some elements in the X-slot (e.g., bare nouns) to what Quine (1969) termed observation sentences, like “Dog!”, where joint attention, or empathy in the Quinean terms, is essential for the understanding thereof. Therefore, indicating closeness or intimacy—and hence inviting the hearer’s empathy—in this manner is a shrewd strategy to get along with others, particularly in the environment of online communication, where nonverbal information such as facial expressions and paralanguage is not available (cf. Naya 2017).²⁴ Given the joint attention effect, example (38b), repeated here as (48), may be further elaborated.

As observed in Section 5.5, the discourse marker you know may function as a private expression, since the speaker can merely “pretend” that the hearer also knows what the speaker has in mind (cf. Swan 2016: 301). Thus, the speaker may use you know without considering the hearer’s actual knowledge state, i.e., whether the hearer really knows the reason or not. However, to the extent that the speaker is conscious of the hearer when using you know, sentence (48) should be considered an extended instance sanctioned by the joint attention effect, which is entrenched in the because X construction. Therefore, in response to Bohmann’s (2021: 33) question whether the because X construction should be treated as one unified category or a category of more or less prototypical and extended instances, this article takes the latter position. Once the construction is entrenched, apparent public expressions may appear in the X-slot, as they are coerced into private expressions by the construction (cf. coercion by override (Audring and Booij 2016)). Combined thus with the construction’s joint attention effect, the meaning of the discourse marker invites the hearer’s attention to the speaker’s thought and hence makes it possible for the speaker to share with the hearer an unstated reason, which may or may not be different from the hearer’s understanding. Because of the private nature of the X-element, the speaker does not have to care about whether he or she can appropriately communicate the intended propositional content with the expression. Another discourse marker reported in Mendes Junior and Mattos (2021) (because see?) may be explained along the same line.

The following examples reported in Bohmann’s report, which contain definite noun phrases, should also be treated as extended instances sanctioned by the construction’s joint attention effect.

As argued in Section 5.2, nouns in the X-slot should be basically bare (McCulloch 2012; Bergs 2018). The definite noun phrases are used in (49a, b); crucially, however, they do not simply refer to bad acting in general and high temperature but are used as euphemisms.²⁵ With the definite article the in (49a) and the demonstrative this in (49b), the speakers indicate that the euphemistic senses are shared with the hearers. Euphemisms shared between the interlocutors in such a way are similar to in-group jargons in that the use of general words are understood in specific senses between the interlocutors. Brown and Levinson (1987: 103ff.) view the speaker’s use of in-group jargons as a positive politeness strategy to claim “common ground” with the hearer and as indicating that “he [= the speaker] wants to ‘come closer’ to H [= the hearer]” (p. 103).

McCulloch’s (2014) example with a definite noun in (50) also exemplifies this characteristic.

Example (50) is based on the well-known English expression the dog ate my homework, a phrase used (humorously) by children to make a poor excuse for their failure to turn in homework assignments. As the stale phrase is well-recognized in the speech community, the speaker can claim common ground and highlight friendliness with the hearer by using the definite noun the dog.

In summary, the expressive structure represented in (16) is entrenched in the construction by the use of its prototypical instances.²⁶ Because of the privateness of the X-element, the construction bears a function to indicate the speaker’s closeness or intimacy to the hearer; once established as such, the construction extends its use and accommodates apparent public expressions in the X-slot (e.g., (48), (49a, b), (50)). Even though the forms of the X-elements indicate the publicness, the expressions are interpreted in accordance with the construction’s function. Put differently, the construction sanctions such apparent public expressions as long as they do not conflict with the constructional function. Cappelle (2021: 28) observes that “in many of the because X cases, the speaker conveys something like ‘I know that I’m presenting the reason in a very condensed, almost cryptic way, but I’m sure you know what I mean by it’” (underline added). That is, using the because X construction, most speakers believe that mutual understandings with the hearers are established about the reason. The discourse marker you know in (48) well represents this characteristic; the definite noun phrases in (49a, b) and (50) indicate that the information is shared between the interlocutors. Although the speaker may condense the information regardless of the hearer’s actual knowledge state, the speaker may also indicate that the condensed information is understood by the hearer. The co-operative hearer has to decode the “cryptogram”; as a result, a joint attention effect is reinforced. As such, just because some X-elements look like public expressions does not mean that the construction extends its use to an unlimited extent. The X-elements sanctioned by the construction are limited to those whereby the speaker can indicate his/her assumption that the condensed information is shared with the hearer.

8. Conclusion

This article has claimed that the X-element in the because X construction represents the speaker’s private expression while the whole construction functions as a public expression. To account for the two-layered expressive structure, using the notion of small discourse, I have argued that the speaker who starts the discourse with the unmarked mode of expression in English switches his/her perspective to the inside perspective as the discourse progresses. Although these basic claims are in line with Kanetani (2017), the present article has emphasized the significance of the two-layered expressive structure and given an answer to the question left open in the earlier work: what effects are brought about by the proposed two-layered expressive structure. By encapsulating private expression within the public expression, the speaker becomes involved in the situation while avoiding the abruptness. At the same time, by using a subjective expression, the speaker also brings the hearer closer to him-/herself (cf. Hirose 2013), so that the hearer feels intimacy toward the speaker (cf. Hasegawa 2010; Naya 2017). The speaker may utilize the constructional function established as such and strategically embed in the X-slot apparent public expressions such as discourse markers (e.g., you know, see?) and definite nouns (e.g. the bad acting, this fever), indicating that he or she assumes that the hearer knows what the speaker means by them. As a result, the hearer is forced to construe these expressions as being shared and to be involved in the speaker’s situation construal.

I conclude this discussion by comparing the present argument with the treatment of the construction in the previous studies. Bohmann (2016) views the densification of information as a motivation (cf. Biber and Finegan 2001), while Okada (2020) characterizes the X-element as a reference point for the conceptually relevant proposition. These observations seem to account for essentially the same mechanism from different directions. As a reference point, the X-element facilitates the comprehension of the target propositional meaning (cf. Langacker 1993). To do so, the proposition is condensed into the most salient element, which may function as a “keyword” that best represents the proposition. The present conclusion that the X-element is subjective enough to attract the hearer closer to the speaker and accordingly bring about a joint attention effect is compatible with the previous views. As noted in Section 7, the X-element is similar to an observation sentence used when the speaker expresses a situation as he/she construes it. When the speaker provides the X-element as a keyword, the hearer empathetically viewing the situation from the situation-internal perspective is ready to recover the intended propositional content based on it. This conclusion also directly supports Bergs’s (2018) observation that the construction is subjective. However, the hearer’s commitment toward what is being said, as well as the speaker’s commitment, is crucially involved.

Data availability

All data underlying the results are available as part of the article and no additional source data are required.