Volume 51, Issue 1-2 - Exclamation, intensity, and emotive assertion

In this paper, we present a novel analysis of exclamations as assertions equivalent to particular uses of declarative sentences with emotive verbs. Focusing on wh -exclamatives and declarative exclamations, we offer a wide range of arguments for why they are both assertive. We further argue that like emotive verbs, exclamations convey a presupposition not of factivity but of subjective veridicality anchored to the speaker, and assert the emotion (of surprise, amazement, or a negative emotion). Our analysis proposes a syntax-semantics for exclamations without a speech act operator, and exclamativity surfaces as an attitude rather than a speech act. This seems to be well motivated by the Greek, German as well as English facts we examine in the paper. Illocutionary operator approaches cannot capture the facts discussed here, and they also fail to determine precisely what the exclamative force might be.

In their target article entitled “Exclamation, intensity, and emotive assertion,” Trotzke and Giannakidou argue that exclamatives such as German Wie schnell Ada war! ‘How fast Ada was!’ are equivalent to emotive assertions such as Ich bin sehr überrascht, wie schnell Ada war ‘I am very surprised at how fast Ada was.’ This commentary challenges this equation on both empirical and theoretical grounds. Focusing on German, I discuss the external combinatorics of exclamatives, their internal make-up, and also the interpretation of the underlying emotive attitude. The discussion draws attention to substantial formal and functional differences between exclamatives and emotive assertions. Furthermore, it is argued that their commonalities should not be attributed to supposedly identical structures, but to independent common properties.

This commentary offers a personal assessment of Trotzke and Giannakidou’s proposal and explores two key aspects: the status of the descriptive meaning of emotive assertions and the emotions conveyed by exclamations. By revisiting the data points and claims presented in the original paper, it raises questions that may contribute to a deeper understanding of exclamations.

This commentary paper is a response to the target article by Andreas Trotzke and Anastasia Giannakidou, who argue that neither exclamatives like How smart Anya is! , nor declarative exclamations like Anya is so smart! have a special, exclamatory illocutionary force, but instead are both semantically equivalent to declaratives that assert propositions about the speaker’s attitude like I am amazed at how smart Anya is with respect to not only their compositional structure, but even what is at-issue and what is presupposed. In this response, I defend a specific version of the more traditional view under which exclamatives express affect about presupposed content, declarative exclamations assert the content of the declarative with expression of affect “on the side”, and neither assert propositions about one’s attitude. I do so by challenging Trotzke and Giannakidou’s claims about the behavior of these different types of sentences in the discourse and by looking at a wider typology of exclamations within and across languages, with special focus on their prosody and more fine-grained aspects of meaning.

This commentary expands and tries to improve on the use of embedded exclamations to clarify what it means for exclamations to be assertive and draw further conclusions on the nature of the subjectivity and emotivity that everyone admits that exclamations convey. Although much attention was paid to emotive embedding, considering its coexistence with epistemic embedding and the actual prevalence thereof clarifies the way the multiple layers of exclamatory utterances interact. In particular, the presence of a non-emotive, bona fide epistemic assertion brings back into play the interlocutor’s subjectivity, which must be inscribed in the meaning of exclamation, along with the speaker’s.

This comment on Trotzke and Giannikadou’s target paper entitled “Exclamation, intensity, and emotive assertion” addresses the question as to how the balance between descriptive information and emotional expressivity in sentences like How hard John works! and John works so hard! is best conceptualized in a proper grammatical-pragmatic framework. Whereas T&G stress the primacy of the descriptive information, the present comment supports an alternative view, in which the expressive aspect of such utterances is ‘upgraded’. In terms of speech acts, T&G consider the utterance of such sentences as assertions, while they are conceived here, in line with tradition, as exclamations. For the proper modelling of the status of the constructions discussed, the importance of a clear distinction between describing and indicating emotions is stressed.

Trotzke and Giannakidou advance the strong hypothesis that exclamative sentences and exclamations can be reduced to a combination of three basic sentence-types (declaratives, interrogatives, and imperatives) with added intensity for emphasis. Moreover, they give convincing arguments for treating root exclamatives as clauses subordinated to an emotive predicate. In this commentary I briefly evaluate such claims on empirical and theoretical grounds.

This article challenges the claim that exclamative sentences lack an illocutionary force of exclamation and are semantically equivalent to emotive assertions. Drawing on German data, we show that there are different classes of exclamatives that vary with respect to deniability and do not function well as answers to information-seeking questions but rather serve as replies in discourse. We argue that all the different classes need to be taken into account to arrive at appropriate generalizations for the phenomena under discussion.

In this response, we clarify some of the main issues that kept surfacing in the seven commentaries. We acknowledge and highlight the significance of the diversity of linguistic means and syntactic configurations that can be used for what has been traditionally called exclamation, and we also address the issue of diverging judgments and introspective data. Finally, we look at the topic from a broader perspective, concluding our discussion with a new proposal for interjections such as ouch! and wow! as super-linguistic in character, a kind of verbal emoji.