Non-standard questions in English, German, and Japanese

1 Introduction

In this paper, I will discuss the nature of non-standard questions (NSQs) in English, German, and Japanese. I will first introduce some basic properties of NSQs and then show that NSQs can be classified into two types: the speaker-oriented type and the addressee-oriented type. I will then suggest that the addressee-oriented NSQs can be further classified into the counter-expectation type and the aggressive question type. I will finally summarize the whole discussion.

2 Basic properties of NSQs: [+speaker-oriented] and [+addressee-oriented]

In this section, I will introduce some basic properties of NSQs by comparing them with standard questions (SQs) with reference to sentence final particles (SFPs) in Japanese. Japanese is very rich in SFPs, which encompass a complicated and diverse range of usages, from typical to atypical, according to context and interpersonal relationships (see Davis 2009; Endo 2007, 2012, 2020; Endo and Haegeman 2019; Kamio 1994; Takiura 2008; Tanaka 2000; Tokieda 1951 for various properties of SFPs in Japanese). Among various properties of SQs and NSQs shown by particles in German and Japanese, I will mainly focus on the information-seeking properties, which will be relevant to the discussion to follow.

Let us first examine the contrast between the information-seeking SQs in German exemplified in (1a), where the speaker expects an answer from the addressee, and the non-information-seeking NSQs in (1b), where in the presence of the discourse particles nur and bloß, the speaker is more interested in expressing his/her emotion than seeking information from the addressee. Bayer and Obenauer (2011: 468) call the NSQ in (1b) ‘Can’t-Find-the-Value-of-x Question (CfvQ).’^[1]

The German NSQs and NSQs in (1a and b) correspond to the Japanese NSQ and the NSQ in (2a and b), where just like the German case we saw immediately above, the bold SFP na yields a NSQ. What distinguishes Japanese NSQs from German NSQs is that Japanese NSQs have a particle in the clause-final position, while German NSQs have a particle in the clause-medial position.

The semantic import of the Japanese SFP na in (2b) is to signal the expression of weak confirmation by the speaker about the proposition based on some evidence. For instance, in the absence of the SFP na, the sentence in (3) is a neutral statement reporting the fact that a house is on fire. On the other hand, in the presence of the SFP na, the sentence implies that the speaker has some evidence for the proposition such as the fire siren. A reviewer asks what is the contribution of the evidential SFP na to the interpretation of NSQ in (2b). To understand this point, consider Maslova’s (2003) view that evidentiality reflects the perspective of the speaker. To be more specific, evidentials in interrogatives, as in declaratives, are understood as giving the speaker’s ‘take’ on the described event, where the speaker is ‘wondering’ about something rather than seeking information from the addressee. Thus, in the NSQ in (2b) the combination of interrogative marking ka and evidential element na indicates that the speaker is ‘wondering’ about the exact location of his/her glasses without seeking information from the addressee. What is crucial here is the fact that in the presence of the SFP na, the sentence in (3) is most naturally interpreted as a monologue without the addressee. The fact that the SFP na is used as a monologue is captured by assuming that it carries the feature [+speaker-oriented].

With this point in mind, let us consider another type of NSQ in German exemplified in (4), i.e., the rhetorical question, which is slightly different from the CfvQ in (1).

This rhetorical question can be most naturally translated into Japanese in (5), where the sentence is suffixed by the SFP yo, and not by na – the SFP we saw above. The SFP yo indicates the speaker’s mental attitude that the stated proposition before the SFP is under the control of the speaker (Takiura 2008). Accordingly, it is typical to use the SFP yo to elicit a sense of insistence or to get the addressee’s attention (e.g., Davis 2009; Kamio 1994). What is crucial about the SFP yo in the present context is that the sentence suffixed by the SFP yo is not used as a monologue. Thus, when the NSQ in (3) is suffixed by the SFP yo, instead of the SFP na, it can no longer be interpreted as the speaker’s monologue but implies the presence of the addressee. In this sense, the SFP yo in (5) can be characterized by the feature [+addressee-oriented]. I will discuss further properties of the SFP yo later on.

So far, we have looked at two types of NSQs in German and Japanese, both of which make use of particles to form NSQs. A note of caution is in order here. Although German and Japanese NSQs are similar in yielding NSQs by making use of particles, particles of these two languages are not entirely identical. As the careful reader might have already realized, the particles in Japanese we saw above can be used in non-interrogative sentences as well, which would not be possible with the particle mentioned above in German.^[5] In this paper, I will not discuss the issue of where this difference between the two languages comes from. Rather, I will concentrate on the common property of these two languages to create NSQs by using particles, especially the property of particles modifying or fine-tuning illocutionary force. To illustrate this point, let us consider the fact, originally pointed out by Jacobs (1986, 1991, that discourse particles in German modify different illocutionary forces like directives in the sense of Searle (1985). For instance, consider the sentences in (6), which are borrowed from Coniglio and Zegrean (2012: 233) (cf. Coniglio 2014 for the discussion of discourse particles in German). Here, the clause type is imperative in all cases, but the pragmatic strength of the order is modified and ordered according to the meaning of particles.

In parallel to this, illocutionary force in Japanese is modified by various types of SFPs, where the pragmatic strength in (7) below is ordered in the same way as the German case above, as shown by Dohi (2020).

3 NSQs in English

Keeping in mind what we saw in the previous section, let us next examine NSQs in English with special attention to negative islands, which prohibit an adjunct wh-element such as how from moving across a negative element such as not. Shlonsky and Soare (2011) note that unlike why, how cannot cross negative islands, as illustrated in (8):^[7]

However, Andrew Radford (pers. comm.) points out that (8b) is acceptable for him in the following kind of context in (9a). This is a kind of rhetorical negative question, found in a discourse like (9b). Note that the exclamation mark after the question mark marks this sentence as a rhetorical question. There is something different about rhetorical negative questions – for example, they do not license polarity items like (partitive) any, as in (10). In addition, they have a different intonation from operator questions.^[8]

Here, we see that NSQs in English are formed in a way different from NSQs in German and Japanese in that English does not make use of particles to yield NSQs. To see this point, let us consider the following how come questions.

Attributing the observation to Andrew Simpson (pers. comm.), Tsai (2008: 89) mentions the difference in meaning between how come questions and why questions: why questions involve no special expectation about whether or not a given state of affairs should hold, whereas how come questions expresses surprise that a particular state of affairs should hold, as illustrated in (11). However, it is not the case that how come always express surprise that a particular state of affairs should hold. For instance, there is no surprise in what B says in (12) below, just curiosity and how come sounds less invasive than why in this context.^[11]

Where does this difference of how come questions come from? To find an answer to this question, it would be helpful to consider the following sentences from the comic Peanuts, where the Japanese sentences corresponding to the two types of how come questions we saw above are clearly expressed by different SFPs.

In (13), Lucy addresses a how come question to Snyder and receives a response with the sentence prefixed by because. Here, the corresponding Japanese how come question in (13b), which is translated by Syunzi Tanigawa, sounds like a SQ suffixed by no SFP, where mild curiosity is felt by Lucy. In contrast, in (14), although Linus uses a how come question with Lucy, he does not receive a response prefixed by because but only a comment from her. The corresponding Japanese how come question sounds like a NSQ with strong irritation expressed by the speaker. In this case, the sentence in Japanese is suffixed by the SFP sa, as in (14b). In the comic, Linus’s face and gesture show stronger emotion in (14) than we see on Lucy’s face in (13) (see Endo 2019 for this point with pictures). At this point, one may naturally wonder what is the SFP sa creating a NSQ with the meaning of surprise in (14). The SFP sa is used to report the speaker’s familiarity with the proposition, implying that the sentence suffixed by this particle should be taken as a matter of course. According to Uyeno (1971), the meaning of the SFP sa is contrasted with the meaning of the SFPs yoo ‘appear,’ rasii ‘seem,’ and soo ‘hear,’ which are used when the speaker’s judgment is made based on appearance. The SFP sa, in contrast, is used when the speaker’s judgment is already made based on his/her own supposition or inner feeling. Because the speaker’s judgment is already made in uttering the SFP sa, the speaker’s supposition is taken to be discourse-familiar, and thus, we cannot start a discourse with a sentence with the SFP sa, as illustrated by the following contrast.^{[13] , [14]}

Based on this fact, I suggest that the SFP sa expresses the speaker’s epistemic judgment. How can we capture this property of the Japanese SFP sa in the meaning of surprise in how come question? Cinque (1999) postulates various functional heads to host mood elements in the clause structure. According to this idea, the particle sa can be thought of occupying the head position of the functional projection of epistemic mood to give rise to express the speaker’s inner emotion of surprise in NSQs. Assuming Chomsky’s (2001: 2) Uniformity Principle to the effect that in the absence of compelling evidence to the contrary, assume language to be uniform, I suggest that how come questions expressing surprise in (14) are formed by covertly activating the functional head of epistemic mood that hosts the Japanese SFP sa to create the meaning of surprise. In this sense, we might consider English as a “silent/telegraphic language” where those functional heads that are utilized to yield NSQs in German and Japanese remain silent or unpronounced.^[15] See Endo (2020) for the mechanism of covertly activating a functional head responsible for a mood element in the framework of the cartography of syntactic structures (Cinque 1999; Rizzi 2014).

4 Counter-expectation

Finally, let us explore the possibility of further splitting addressee-oriented NSQs into several types by looking at another SFP no, which expresses the speaker’s counter-expectation in the matrix clause. To see what the speaker’s counter-expectation looks like, consider the following sentences in (16a and b). In the absence of the SFP no, the speaker asks the addressee if he/she would like to eat what the speaker recommends. In contrast, in the presence of the SFP no, the speaker sounds more interested in expressing his/her surprise than seeking information from the addressee while witnessing the addressee about to eat what the speaker does not expect him/her to eat. In this sense, the SFP no can be thought of carrying the feature [+counter-expectation] to form NSQs.^[16] A reviewer asks what is the exact difference between “inner emotion of surprise” expressed by the SFP sa we saw in the previous section and “counter-expectation” expressed by the SFP no we are looking at here. One of the most distinguishing properties between these two SFPs lies in the fact that the SFP no, not the SFP sa, serves as the licensor of wh-expressions asking for reason such as naze ‘why’ in Japanese. To see this point, consider the sentences in (16c–e) below. In colloquial Japanese, interrogative sentences can be formed simply by having a final rising intonation as in (16c and d) except for why-questions as in (16e), which always require the SFP no as in (16f). Interestingly, when the SFP no is suffixed to a non-why question as in (16b), the sentence always carries the meaning of what might be called “concealed why-questions” in the sense that the sentence is interpreted as why is it that …?, especially in the matrix clause.^[17] Thus, the accurate translation of the sentence in (16b) is Why is it that you eat this, where the speaker sounds confused or irritated without seeing any reasons for the addressee to eat what the speaker does not expect him/her to.^[18] Based on this fact, we can reasonably assume that “counter-expectation” expressed by the SFP no is quite different from “inner emotion of surprise” expressed by the SFP sa, which has no concealed why-question interpretation.^[19]

This particle no can also be suffixed by the addressee-oriented particle yo that we saw earlier, as in (17) below. Here, in the presence of the two particles no and yo another type of addressee-oriented NSQ is formed, where the speaker expresses aggressive attitude toward the addressee, which I call the aggressive-attitude type.

In order to see what aggressive attitude looks like, let us consider the following sentences uttered by Lucy in the comic Peanuts:

In (18), Lucy is asking a favor of Charlie Brown; she seems to be able to read Charlie’s mind. In the Japanese translation in (18a′), the sentence is suffixed by the particle ne carrying the feature [+empathy], which is related to mind reading (see Kamio (1994, 2002 for the discussion of empathy revolving around the particle ne). In contrast, in (19) Lucy gives Snoopy an angry warning, where she does not seem to have empathy towards Snoopy because when a speaker gives an angry warning to an addressee, s(he) would be more concerned with expressing his/her emotion than worrying about how the addressee would feel as a result of his/her warning and anger (see Endo 2019 for the exposition and pictures). The corresponding Japanese sentence is suffixed by the SFP yo without the feature [+empathy]. Given this, the meaning of aggressive attitude expressed by the particles no and yo in (17) can be attributed to the combination of the feature [−empathy] carried by the SFP yo and the feature [+counter-expectation] carried by the SFP no.^[21]

5 Conclusion

To conclude, I have discussed the nature of NSQs in English and German by comparing them with Japanese with special attention to various types of particles. I showed that NSQs can be classified into the speaker-oriented type and the addressee-oriented type. I further suggested that the addressee-oriented NSQs can be further divided into the counter-expectation type and the aggressive attitude type by looking at the meaning of particles in Japanese.