Exclamatives, try to be a bit more assertive!

1.1 Advantages

One case where the proposal by Trotzke and Giannakidou can be of much profit is that of the infamous “exclamation intonation” (Burnett 2010; Rett and Sturman 2021), a kind of magic wand that converts the sentence it touches into an exclamation. As Huddleston and Pullum (2017: 923) remark “The concept of exclamation is, moreover, a somewhat nebulous one, and it is not possible to present a well-defined set of grammatical constructions that express exclamatory meaning; very often, of course, it is signalled prosodically rather than, or as well as, by the lexicogrammatical form” [my emphasis, XV]. Even though the authors explicitly leave intonation aside, one can easily imagine that this intonation is a prosodic correlate of the “attitude of emotion” that can be assigned to the question and imperative exclamations in Trotzke and Giannakidou’s (64b) and (64c), respectively. Moreover, the approach can easily explain the fact that in languages where exclamative sentences are typically associated with a particular marker, like Mandarin Chinese, the presence of the marker is not a sufficient condition for being interpreted as exclamative, and the marker may appear in interrogative sentences as well (Badan and Cheng 2015):

While these facts have been a nightmare for any attempt to build a universal categorization of the exclamative sentence-type, they follow easily from the proposal by Trotzke and Giannakidou.

Another similar case where the target paper may offer us new solutions is covert exclamatives in Spanish (Masullo 2017). This construction is formally a declarative sentence with a suspended final contour, but it is interpreted as a degree wh-exclamative, as Masullo’s paraphrases show:

This case will receive a straightforward analysis under the proposal in the target article: we are dealing with wholesale assertions encoded by declarative sentences, but the addition of an emotion intensity yields an exclamation interpretation.

These positive notes also extend to a second major claim of the target article (p. 6):

Our claim is that (3) [= How fast Eliud Kipchoge was!] and (4) [= I am amazed at how fast Eliud Kipchoge was!] are identical in terms of what they assert and presuppose, and we will group them together under the label ‘emotive assertions’. They both assert that the speaker has the emotion of amazement towards the believed proposition that ‘Eliud Kipchoge was extremely fast’, and presuppose that the speaker has the belief that Eliud Kipchoge was extremely fast.

Assuming that How fast Eliud Kipchoge was! is asserting an emotion is not obvious and raises important concerns. Yet, I would like to suggest that this idea might be very helpful for explaining the behavior of tag questions in exclamatives. As remarked by Huddleston and Pullum (2017: 922), exclamatives plainly accept tags with reversed polarity, just as declaratives, but are less natural with constant tags:

Interestingly, they explain the contrast in terms of the kind of content we are seeking agreement on. Whereas the reversed polarity tag is typically suited for seeking agreement on the speaker’s attitude (“Wasn’t it noteworthy/surprising?”), the constant tag seeks agreement on the presupposed proposition (“Was it a disaster?”), which is clearly infelicitous. Hence, this would confirm the claim that we do have the assertion of an emotion.

1.2 Some pending issues

I have shown that the proposal of dissolving exclamations into assertions has an undeniable potential to solve long-standing problems and opening new avenues for research. Nonetheless, as happens with bold and interesting hypotheses, it must face formidable empirical challenges as well. Obviously, to be fair, one cannot expect all problems to be accounted for in a single programmatic article, so my intention here is to highlight some theoretical and empirical aspects of the analysis by Trotzke and Giannakidou that should be given a proper answer in the future.

Let me begin with a victim of their analysis: Ouch! and their siblings. The path taken puts I am in pain. (1b), I am in pain! (1c) and What a pain! (1d) into the box of assertions, so we can dispense with exclamations as a speech act (“exclamation characterizes assertions, but also questions and imperatives; it is therefore not a distinct speech act” p. 45). While this might be certainly advantageous on several grounds, as I have suggested above, it leaves us without a proper account for purely expressive acts like Ouch! or Wow! I don’t think that anyone will be happy saying that these utterances are assertions of the speaker’s emotion, for there is no propositional content we can resume with an anaphor:

Yet, if they are not assertions, what kind of beast are they? In classical speech act terms (see also Unger 2019), these utterances do not seem to have a direction of fit (the hallmark of expressive speech acts), unlike assertions, imperatives or questions. Hence, if we do not have an exclamation speech act, these small particles might be the elephant in the room for the proposal by Trotzke and Giannakidou.

A second problem concerns the idea that exclamative sentences like How fast Eliud Kipchoge was! are asserting the speaker’s emotion, just as I am amazed at how fast Eliud Kipchoge was. If this emotion were asserted, it should be accessible to event or propositional anaphora. However, whereas it is clearly available for anaphoric resumption in the declarative in (6), it is clearly not in the exclamative in (7):

Crucially, (7B)–(7B′) can only mean “I am very fast too” and “it is obvious that Eliud Kipchoge was very fast”, respectively. This contrast suggests that the propositional content “I am amazed that p” is not accessible in (7), which is certainly unexpected under the assumption that both utterances assert the emotion of the speaker.

As far as I can see, this is not surprising. While it is true that exclamations can be used as indirect responses about the speaker’s emotional state, given the proper information structure pattern, they are not used as direct answers, unlike assertions. The correlation uncovered by Trotzke and Villalba (2020) was that “Polar questions can be followed by that-exclamatives, and non-polar questions can be followed by both wh-exclamatives and declarative exclamations.” However, corpus examples are hard to find, and even dialogs fitting into the required pattern are deemed impossible:

These examples are not intended to deny that exclamations can be used as responses in some contexts, but they clearly suggest that, if they are just assertions with intensity, we fall short in offering a proper explanation of their poor behavior in information seeking contexts.

Indeed, my impression is that these uses as responses are not particular about exclamatives or exclamations. We know that interrogative sentences are also fine as indirect responses. The prototypical case is rhetorical questions (Biezma and Rawlins 2017; Caponigro and Sprouse 2007; Sadock 1974), as in the following examples from Biezma and Rawlins (2017):

These rhetorical questions do not expect an answer (albeit they may have one) and are equivalent to assertions. In the literature (e.g. Caponigro and Sprouse 2007; Biezma and Rawlins 2017), it is argued that we do have a real question, but in the context at hand, its contribution indirectly updates the common ground as an assertion would do directly. In other words, we don’t need to collapse both speech acts into one another to explain their similar context-change potential. This rationale can be applied to our exclamatives as responses: they simply make a similar contribution to the common ground (or the question under discussion, if you prefer) as an assertion, but in an indirect way, as Trotzke and Villalba (2020) remark. Obviously, this extends to all cases of indirect speech acts: if declaratives, interrogatives or imperatives may encode different speech acts, it is not surprising that exclamatives may too.

As I have remarked above, by raising these issues I am not trying to invalidate the proposal by Trotzke and Giannakidou to analyze exclamatives as assertions, but I would like to suggest that the hypothesis requires some refinements, and a wider set of evidence is needed to evaluate it properly.