Exclamatives: a grammatical category?

Els Elffers
University of Amsterdam

1. Introduction

In most Western European grammars, sentences such as Hurrah!, How very curious!, or Vienna is so dull! are categorized as exclamatory sentences or exclamatives. Next to declaratives, interrogatives and imperatives, exclamatives are usually regarded as a separate sentence type.

However, as a grammatical category, exclamatives are more problematic than other sentence types. More often than other sentence types, exclamatives are omitted from grammars, or they are dealt with very succinctly, and/or in a rather ambiguous way.

During the last decades, there has been a cry for more research into exclamatives. This is mainly due to a growth of interest in themes such as “language and emotion” and “the expressive function of language” (cf. e.g. Foolen 1997). Below, I will briefly discuss the history of thought about exclamatives. Special attention will be paid to some early insights into the problematic character of the category. I will argue that, despite some theoretical improvements, the category has remained problematic up until the present day. Solutions are within reach only if two long-standing ideas are given up: (i) the idea that exclamatives constitute an independent category, (ii) the idea that research of exclamatives exclusively belongs to the “language and emotion” area.

2. Where do exclamatives come from?

Exclamatives are not among the categories that can be found in the earliest Latin-oriented grammars of European vernaculars. In these 16th and 17th century grammars, functional sentence typologies are still lacking. However, the very idea of sentence functions was already present, if only through the general adoption of Latin modi of the verb (indicative, conjunctive, imperative). Moreover, some notion of exclamatives must have been present, because all grammars of those days deal with spelling and punctuation. Among the punctuation marks, the exclamation mark is discussed. For example, the 17th century Dutch grammarian Leupenius claims that this wonder-teken (his Dutch equivalent of Latin nota admirationis, “mark of surprise”) has to be added after an exclamation. His 18th century colleague Moonen applies the term “mark of complaint, joy or surprise”. He distinguishes the exclamation mark and the question mark from other punctuation marks, because they imply that sentences are pronounced “in a different tone”.[1]

Only in the course of the 19th century did functional sentence typology become a separate subject of systematic linguistic reflection. This was mainly due to the rise of psychological linguistics, especially in Germany. The earlier view of the prototypical sentence as a the two-membered embodiment of a logical subject-predicate judgment was gradually abandoned. Sentences with quite different functions and quite different forms (especially many kinds of one-membered sentences) became focuses of interest. A question was raised, which, actually, has never disappeared from the linguistic agenda: how many and which functional sentence types should be distinguished?

In this discussion, exclamatives played their part from the very beginning, but this part was soon becoming controversial. Already around 1900, several linguists noticed the category’s problematic character; some of them refused to recognize exclamatives as a separate category. Despite these troubles, the category has been vindicated until the present day. For example, Quirk et al. (1985) distinguish exclamatory sentences on a par with declaratives, interrogatives and imperatives.

3. Problems with exclamatives

Which properties of exclamatives were thought problematic? Formal as well as semantic properties appear to play a crucial role. In the first place, there were demarcation problems: boundaries with other sentence types turned out to be unclear. I will discuss some insights into these problems in the work of, respectively, the Dutch grammarian Cornelis den Hertog (1846-1902) and the well-known German linguist Hermann Paul (1846-1921). In the second place, there were conceptual problems with the alleged “emotion-expressive” character of exclamatives. Early signals of these problems are presented by the German linguist Georg von der Gabelentz (1840-1993). I will discuss his ideas in combination with later elaborations of the issue by the Hungarian-Dutch psychologist Géza Révész (1878-1955).

3.1 Demarcation problems
3.1.1. Problems of formal demarcation. Den Hertog

In Western European languages, specific syntactic constructions can be found that are characteristic for exclamatives. For example, in English, there are exclamative constructions such as in (1)-(3), with exclamative pronouns/adverbs or with a subordinate clause structure used as a main clause:

  • (1) What a devil of a name! What a fool I was!
  • (2) How very curious! How fast she can run!
  • (3) That he should have left without me!

Many sentences consisting of an interjection are also typically exclamative, e.g. Hurrah!, Dammit! On the other hand, however, nearly all other syntactic sentence types can be used in exclamative sentences. (4)-(6) exemplify the exclamative use of, respectively, the declarative, the interrogative and the imperative type:

  • (4) He was such a terrible referee!
  • (5) Isn’t Christine clever!
  • (6) Sit down!

The existence of examples such as (4)-(6) has motivated some grammarians to deny the existence of a separate exclamative sentence type. Den Hertog, in his well-known grammar of Dutch (Nederlandsche Spraakkunst 1892-1896), answers to a fictional opponent who asks why he does not distinguish exclamatives: “The answer is that this would be necessary only if language would possess specific forms for such utterances. Now, this is not the case. […] An exclamation can be regarded as a special type of declarative: The Lord is great!, which can also take the shape of an interrogative: Would he do a thing like that? (Den Hertog 19032: 16, my transl. E.E.).

3.1.2. Problems of semantic demarcation. Paul

The general view of the function of exclamatives was always that they express the speaker’s emotion. The linguist-psychologist’s Wundt’s (1832-1920) statement “Der Ausrufungssatz ist als solcher Ausdruck eines Affekts in sprachlicher Form” is a clear expression of this view (Wundt 19224: 259).

Paul, however, argued against Wundt, claiming that his approach made possible the incorporation of any sentence in the class of exclamatives. Therefore: “Der Anteil des Gefühles kann keinen Einteilungsgrund für die Satzarten geben. Er kann bei jeden Satzart stark oder schwächer oder gar nicht vorhanden sein” (Paul 19205: 133). Paul therefore rejects the “Ausrufungssatz” as a grammatical category.

Formal and semantic aspects of exclamatives seem to suffer from similar inadequacies: they are unable to demarcate a restricted category of sentences as exclamatives. Den Hertog’s and Paul’s views are each other’s mirror. Both reflect the basic insight that the exclamation mark, bearing its characteristic meaning, can be put after any sentence.

3.2. Conceptual problems

The assumption that the characteristic meaning of exclamatives is “expression of the speaker’s emotion” has been challenged in various ways. In the first place, the idea of emotion expression, which is mostly conceived as an involuntary behaviour-accompanying phenomenon, seems rather implausible for many exclamatives, and the idea is, moreover, incompatible with the intentional and hearer-directed conception of language use, which gradually became more and more accepted from ±1890 onwards.[2] In the second place, apart from many interjections, exclamative sentences have a propositional content that is conveyed to the hearer as well, in addition to the alleged emotion, cf. The Lord is great! We won the game!

For scholars like Wundt and Paul, the first problem does not yet exist. They still adhere to the general view that any sentence mirrors the speaker’s psychical occurrences and that these occurrences (representations, associations, apperceptions, emotions) constitute the sentence’s meaning. The second problem is clearly observed by Wundt (be it in a psychologized form). His – somewhat artificial – solution is that, in exclamatives, “wengleich auch hier jedes Wort im allgemeinen Ausdruck einer Vorstellung bleibt, doch diese nur als Erregung jener subjektiven Gemütsbewegungen wirkt” (Wundt 19224: 261).

Von der Gabelentz, a linguist contemporary of Wundt and Paul, but in advance of his time in many respects, observed both problems but could not solve them. Révész, a linguistically oriented psychologist, did propose a solution, which is, however, extreme in its denial that exclamatives belong to human language.

3.2.1 Gabelentz: confusion

For Gabelentz (19012: 317-324), exclamatives are exceptions to the general rule that speaking involves an appeal to the listener. This appeal can take the shape of a statement, a question, or a command/request. Exclamations are mentioned as the fourth category, although they lack hearer-directedness. They are uttered when speakers are under the influence of powerful feelings. Such feelings may result in crying, laughing, applauding etc., but also in exclamative language. For Gabelentz, the other three sentence types all represent subcategories of the “mittheilende Gedanke im weiteren Sinne”. Exclamatives constitute a separate category, with its own subcategories. During discussion of formal subcategories, however, boundaries are becoming fluid, because several formal types of exclamatives entirely lack the alleged absence of hearer-directedness, e.g. Brav gemacht! Käme er doch! Exclamatory imperatives, e.g. Halt!, Hut ab! and vocatives such as Polizei! are, for this very reason, classified, not as Ausrufe, but as Zurufe, and belong, therefore, to the “Mittheilungen im weiteren Sinne”. So exclamatives no longer constitute a homogeneous category.

When discussing semantic subcategories, Gabelentz goes even further in bridging the gap with other sentence types. He admits that exclamations not only express an emotion, but also its “Grund”, the propositional content. This “Grund” can take the shape of all varieties of “Mittheilungen im weiteren Sinne”, so that the two categorization schemes, initially sharply separated, turn out to overlap. Gabelentz’s honest conclusion is that, in this area, “die Kunst der Classification” is doomed to failure (Gabelentz 19012: 324).

3.2.2. Révész: (partial) banishment to extralinguistic areas

In Révész (1946), the central theme is the origin of language, and hearer-directedness is argued to be the essential feature of human language. Radicalizing recent psychological-linguistic ideas, especially those of Karl Bühler (1879-1963), Révész rejects all theories that assume a direct development of human language out of pre-linguistic expressive cries, with interjections as the first (proto-)linguistic elements. As Révész still adopts the view that primary interjections and other exclamatory expressions are purely expressive, he has to regard them as non-linguistic. He vaguely mitigates his view of interjections by saying that it does not take into account their “sekundäre Bedeutung als Sprachmittel” and their “Eingliederung in die Sprachen” (Révész 1946: 42).

What about other types of exclamation? Quite consistently, Révész does not recognize exclamatives as a functional category, next to declaratives, interrogatives and imperatives: “Der Ausruf [..]gehört […] nicht zu den Hauptfunktionen der gegenseitigen sprachlichen Verständigung. Sie stellen sprachlich symbolisierte Ausdrucksbewegungen dar, stehen mithin mit den Interjektionen auf einer Stufe.” However, not all exclamations are of this type. Others, e.g. Unerhört!, Weh mir!, Reizend!, Ach! Hallo! Hopp!, “sind bei näherer Prüfung ungezwungen einer der genannten Funktionen, meistens dem Indikativ, zuzuordnen wodurch der Indikativ einen imperativen, optativen oder interrogativen Einschlag erhält” (Révész 1946: 152). As in Gabelentz’s exposition, the category of exclamations, initially sharply demarcated, is actually broken up.

4. Later developments

The above historiographic fragments lay bare the totality of problems with exclamatives as a category. As far as I can see, these problems are still largely unsolved. As to formal demarcation, however, considerable progress was made due to the rise of intonation research, during the first half of the 20th century. Den Hertog’s futile search for formal criteria got a new chance. Actually, Moonen’s – still vague – “different tone” was now regarded as the main criterion for characterizing the various sentence types, more powerful than morphological or syntactic criteria.[3] However, especially with respect to exclamatives, ideas about a characteristic intonation pattern turned out to be largely inadequate.[4]

Only recently, Bolinger (1989: 248-249) argued that this is not surprising, because search for one exclamatory intonation pattern is useless: “Exclamations draw impartially upon the full repertory of up-down patterns. […] What characterizes the class is not shape but range: exclamations reach for the extreme – usually higher, but sometimes lower”.

This idea of “enlargement” of existing patterns appears to be a plausible alternative for the one-pattern approach.[5] Moreover, it shows us a road out of other problems, a road not taken by Bolinger and his followers, by the way. If exclamatory sentences exhibit intonation patterns of all sentences types, be it “enlarged”, why wouldn’t it be the same with the semantics of exclamatives? This would imply that exclamatives are not a separate sentence type on a par with other types: exclamatives are declaratives, interrogatives or imperatives, be it with an “enlarged” meaning.[6] This does justice to Den Hertog’s and Paul’s observation that all sentences can be used as exclamations, but without the category being cancelled. Nor is it retained, in a contradictory way, next to declaratives etc.[7] It also does justice to Gabelentz’s observation that many exclamatives convey a propositional content, just like other sentence types. This is no longer a problem, if the exclamatory meaning is additional to the other meanings. In the same vein, Gabelentz’s and Révész’s problems resulting from the assumption of a strict emotion-expressive meaning of exclamatives disappear. Exclamatives are no longer merely emotion-expressive; they are hearer-directed declaratives, questions, and commands, with an additional emotional accent.[8]

5. Exclamatives without emotion?

However, do exclamatives indeed express the speaker’s emotion? Up to the present day this is the common belief. Recent grammars such as Quirk et al. (1985) and ANS (1997) adhere to this assumption. It can be found in specialist publications such as Bolinger (1989) and Beijer (2002). It is also presupposed in recent publications about “language and emotion” such as Foolen (1997). Only once, I met with an alternative semantic characterization of exclamatives. Klooster (2001:113, my transl. E.E.) claims that the meaning of exclamatives is “vigorously underlining the message. The function of exclamatives is intensifying”.

I am convinced that the latter alternative is the better one. Not only is it more in line with the “enlargement” idea – there is not an addition of a new element, but an intensification/enlargement of what was already there: declaration, question or command – it is also empirically more adequate. This is not the place to defend this idea extensively; I must restrict myself to some suggestive examples. In exclamatives such as Don’t forget your bag! or You may leave the hospital today! the function of the exclamatory aspect (the enlarged intonation pattern) is the enlarged appeal to the listener in terms of illocutionary force: the command/warning and the declarative message are marked as highly relevant. The emotion felt by the speaker, e.g., respectively, a fellow bus passenger and a hospital physician, is, if existent at all, not conveyed. Of course, examples can be found (e.g. We won!) in which the listener “hears” the speaker’s emotion, in this case joy. However, given the earlier examples, I feel that this is an incidental pragmatic implicature, rather than an inherent meaning of exclamatives in general.


[1] Cf. Schaars (1988: 125-129) and Ruysendaal (1989: 203).

[2] This fundamental conceptual change, which regarded psychology as well as linguistics, is discussed in Knobloch (1988: ch.4), Nerlich & Clarke (1996: 4.4) and Elffers (1999).

[3] The observation that sentences such as We won?, with a declarative syntax and an interrogative intonation pattern, function as interrogatives, was made earlier (e.g. by Den Hertog 19032). Theoretical accounts were developed only in the 20th century. For example, De Groot ([1965]2: 49-50) argues that the “subjective layer” (of intonation) always dominates the “objective layer”(of syntax).

[4] For example, De Groot’s ([1965]2: 39) view that the exclamative pattern has only one tonal peak proves incorrect for longer exclamative sentences (e.g. Beijer’s (2002) example What a fool I was not to think of it before!). In the Dutch grammar ANS (1997), exclamatives are not characterized in terms of a specific tone pattern but in terms of a high volume. This idea can be refuted easily. Non-exclamatives may be loud due to e.g. deafness of the listener; exclamatives may uttered very softly, e.g. when the listener is nearby and the utterance is not meant to be heard by other people (e.g. Behave yourself!, uttered by a parent during a reception).

[5] Bolinger’s idea was adopted by other researchers, cf. Beijer (2002).

[6] This idea is very much in line with the general assumption that intonation is an iconic phenomenon.

[7] Quirk et al. (1985) and ANS (1997) are grammars that exhibit this contradictory element by combining the assumptions (i) “exclamatives are a separate sentence type” and (2) “all sentences (declarative, interrogatives, commands) can be used as exclamatives (retaining their declarative etc. meaning).

[8] My view implies that sentences exhibiting one of the exclusively exclamative constructions, such as What a fool I was!, also belong to one of the other sentence types (in this example: declarative). I assume that, in cases like this, the non-existence of a non-exclamatory counterpart has to be explained in terms of the lexical and syntactic content of such sentences, which does not allow a neutral interpretation. This idea has to be elaborated further.


ANS 1997: Haeseryn, Walter e.a. 19972. Algemene Nederlandse spraakkunst. Groningen / Deurne: Nijhoff / Plantijn.

Beijer, Fabian. 2002. The syntax and pragmatics of exclamations and other expressive/emotional utterances. Lund University: Working Papers Linguistics 2.

Bolinger, Dwight. 1989. Intonation and its uses. Melody in grammar and discourse. London etc: Arnold.

Elffers, Els. 1999. ‘Psychological linguistics’. In: Geschichte der Sprachtheorie 4. Sprachtheorien der Neuzeit. Ed. by Peter Schmitter. Tübingen: Narr, 301- 341.

Foolen, Ad. 1997. ‘The expressive function of language. Towards a cognitive semantic approach’. In: Suzanne Niemeier & René Dirven (eds.) The language of emotions: conceptualization, expression, and theoretical foundation, 15-34. Amsterdam / Philadelphia: Benjamins.

Gabelentz, Georg van der. 19012. Die Sprachwissenschaft, ihre Aufgaben, Methoden und bisherigen Ergebnisse. (Reprint, with an introduction by E. Coseriu, 19843). Tübingen: Narr.

Groot, A.W. de [1965]2 [1949] Structurele Syntaxis. Den Haag: Servire.

Hertog, Cornelis H. den. 19032 [1892-1896]. Nederlandsche Spraakkunst. Vol.1. Amsterdam: Versluys.

Klooster, Wim. 2001. Grammatica van het hedendaags Nederlands. Een volledig overzicht. Den Haag: SDU.

Nerlich, Brigitte & David D. Clarke. 1996. Language, Action and Context. The early history of pragmatics in Europe and America, 1780-1930. Amsterdam etc.: Benjamins.

Paul, Hermann. 19205[1880]. Prinzipien der Sprachgeschichte. Halle a. S.: Niemeyer.

Quirk, R. et al. 1985. A comprehensive grammar of the English language. London & New York: Longman

Révész, G. 1946. Ursprung und Vorgeschichte der Sprache. Bern: Francke.

Ruysendaal, Els. 1989. Terminografische index op de oudste Nederlandse grammaticale werken. Amsterdam: Stichting Neerlandistiek VU.

Schaars, Frans A.M. 1988. De Nederduitsche Spraakkunst (1706) van Arnold Moonen. Wijhe: Quarto.

Wundt, Wilhelm19224. Völkerpsychologie. Eine Untersuchung der Entwicklungsgesetze von Sprache, Mythus und Sitte. Zweiter Band, Zweiter Teil, Die Sprache. Leipzig: Kröner.

How to cite this post:

Elffers, Els. 2014. Exclamatives: a grammatical category? History and Philosophy of the Language Sciences. https://hiphilangsci.net/2014/05/07/exclamatives-a-grammatical-category

Tagged with: , , , , , ,
Posted in 19th century, 20th century, History, Linguistics, Syntax, Uncategorized

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: