Navigation – Plan du site

AccueilNuméros12Approches rhétorique, grammatical...Grammar and feelings: a study of ...

Approches rhétorique, grammaticale, phonologique et cognitive du style

Grammar and feelings: a study of Wh-exclamatives in Katherine Mansfield’s short stories

Julie Neveux
p. 193-222


Afin de mieux comprendre le rapport, parfois conflictuel, entre langage et sentiment, j’étudie dans cet article toutes les constructions exclamatives en WH- présentes dans deux collections de nouvelles écrites par Katherine Mansfield, Bliss: and Other Stories (1923) and The Garden Party: and Other Stories (1923).

À la lumière des analyses de mes données, je m’efforce de remettre en question la conceptualisation très répandue des exclamatives comme exprimant évaluation scalaire et surprise suite à la violation d’une attente.

Les exclamatives en HOW sont des énoncés orientés vers le sujet parlant, qui expriment plutôt un sentiment intense dont il ou elle fait l’expérience; les sujets de ces constructions peuvent être omis en raison de leur évidence contextuelle. L’ellipse fréquente du sujet et du verbe résulte donc de l’implication cognitive de l’énonciateur/trice. Dans les exclamatives en HOW, l’énonciateur tente d’identifier un quale évident et saillant en contexte, évidence qui apparaît paradoxalement dans le marqueur d’indétermination WH-; tandis que les exclamatives en WHAT expriment un jugement social et intersubjectif. L’expressivité se concentre dans le syntagme initial en WH-, qui se prolonge en simple expression du sentiment lorsqu’il devient proposition complète.

Haut de page

Texte intégral


Emotion and indeterminacy in Mansfield

1Feelings are difficult to render into language. The more intense they are, the more likely they are to interrupt any form of articulate expression. Emotionally involved speakers might indeed temporarily lose their speaking ability, which requires analytical and constructional skills. Many characters of Katherine Mansfield’s short stories thus experience excessive, physically overwhelming feelings they are unable to express and that further isolate them. Even the first-person narrator of "Je ne parle pas français", whose work it is to write, to select the adequate words, feels speechless when it comes to "meeting" such excessive emotions:

“Good God! Am I capable of feeling as strongly as that? But I was absolutely unconscious! I hadn't a phrase to meet it with! I was overcome! I was swept off my feet!” ("Je ne parle pas français", Bliss, 77)

2The natural, immediate expression of such feelings is indeed a physical one. Trying to impose a linguistic medium on a primarily bodily experience might even alter its essence. Can emotions exist linguistically? Oral expression, be it noise, interjections or cries, might retain the essential physicality of the feelings: one may hear joy in laughter, misery in sobs. But is there any more articulate, oral or written way to make them exist? Mansfieldian characters, when trying to identify their feelings, struggle with words and eventually resort to indeterminate markers such as indefinite pronouns something, somewhere or neutral pronoun It. Josephine, in "The Daughters of the Late Colonel", has lost her authoritative and repressive father a week ago, and does not dare to consciously acknowledge in herself a rising feeling of relief and a longing for levity:

Some little sparrows, young sparrows they sounded, chirped on the window-ledge. "Yeep–eyeep—yeep." But Josephine felt they were not sparrows, not on the window-ledge. It was inside her, that queer little crying noise. "Yeep–eyeep—yeep." Ah, what was it crying, so weak and forlorn?

("The Daughters of the Late Colonel", Garden Party, 125)

  • 1 Onomatopoeia, often used by Mansfield, are described as “mimetic expressivity” by Legallois and F (...)

3Do the sparrows even exist? Are they the fantasied embodiment of the protagonist’s emotion? Such ambiguity, together with the use of the adjective ‘queer’ and the recurrent subject ‘it’, convey Josephine’s difficulty to come to terms with her feeling, whose physicality can be heard in the onomatopoeia1 ‘Yeep–yeep—yeep’. It is again with lexical and grammatical markers of indeterminacy (‘curious’ and ‘somewhere’) that the young Phyllis tries to express her sadness in ‘In Summer’: ‘Oh, I have never been so unhappy before. […] I have a curious pain somewhere’. As Hammond (2016, 56) writes,

To read Mansfield is to reckon with such ambiguously embodied feelings – life for her characters is the experience of obtrusive and often unarticulated emotions. In her most memorable characters we observe emotion in the body: Ma Parker tries desperately to hold back her ‘proper cry’[…], Bertha Young has uncontrollable urges ‘to run instead of walk’ in her moments of bliss […], and the anxious Kezia tiptoes out of ‘Prelude’ feeling ‘hot all over’.

Expressing or describing feeling

4Mansfield’s (often female) protagonists, those characters through whose perspective some parts of the narrative are given in free indirect discourse, are frequently endowed with an acute sensitivity of which they are aware and sometimes ashamed. But this sensibility is also an opening of the self which can be wished for, as Mansfield explains in her Notebooks (1997, 204), for it triggers a "moment which, after all, we live for, the moment of direct feeling when we are most ourselves and least personal". Such a sensitivity has nothing to do with the display of sentimentality which, in Mansfield’s eyes, had so far ensnared female artists, preventing them from becoming good writers. Feeling, in the New Zealand writer’s work, is as powerful as it is intimate. Instead of expanding any egotistic feature of theirs, it enables her protagonists to open up to empathy, a specifically Modernist feeling according to Hammond (2014), and to overcome social discrepancies, thus creating a breath of life and movement in the narrative. This movement is first a reaction, where the self is destabilized. It is a transgressive and disruptive force, which has to be visible as such in discourse, this effort being essential in Mansfield’s motivation as a writer:

For Mansfield, manifest character emotion made literature valuable. In her letters, notebooks and published reviews, she often criticised modernist peers who, in her estimation, wasted their craftsmanship on characters with no apparent emotional life. Dorothy Richardson and Edith Wharton both received this charge from Mansfield, as did E. M. Forster, of whose Howards End she wrote, ‘And I can never be perfectly certain whether Helen was got with child by Leonard Bast or by his fatal forgotten umbrella’. (Hammond 2016, 56)

  • 2 Celle and Lansari (2017, 241) show that surprise is not a prototypical emotion in that "it entail (...)

5In order to keep away from sentimentality, one has to express feeling rather than describe it. This only will preserve the specificity of such intense, indescribable cognitive experience. I have elsewhere (Neveux 2013) distinguished between direct, explicit lyricism (where lexical markers describe emotion) and implicit, indirect lyricism (where syntagmatic relationships as in metaphors and -ness nominalisation communicate emotion). In this paper, I also mainly focus on expression of feeling, although this distinction might not be so clear-cut in the case of clausal exclamatives, the latter end of an exclamative clause having to do with a more descriptive, analytical function. In their study of surprise, in which they also draw on such a distinction between expression and description of this ‘feeling’2, Celle and Lansari (2017, 4) note that

the description of surprise is associated with surprise lexemes and figurative language whereas the expression of surprise is linked to disfluency markers such as exclamations, interjections, etc. This gives credence to a clear-cut division between markers describing surprise and those expressing surprise. Interestingly, it is also suggested that this dichotomy is sometimes insufficient to analyse the aforementioned pragmatic strategies: reenactment and the staging of surprise may well be better captured in terms of expressivity.

6I have chosen to study Wh-exclamatives in Mansfield’s short stories for two reasons. First, they are a highly recurrent stylistic feature of hers in contexts of utterance where intense feelings prevail, maybe the most visible one. Second, they register the conflict above-mentioned, that between language and feeling. They represent, when the exclamative clause is complete, which is not always the case as we will see, a fully-fledged syntactic expression of emotion, all the while still conveying indeterminacy, notably in the Wh-operator but also in other aspects to be described. Such indeterminacy renders the cognitive intensity felt by the speaker, enacting a brief cognitive conflict between feeling and language.

7Celle and Lansari (2015, 87-88), following Michaelis (2001) and Rett (2011), associate Wh-exclamatives with an expression of surprise often denoting speaker-oriented emotional or ‘pathetic expressivity’ (in Legallois & François’s terms 2012). They also mention some cases of more ‘ethical’, co-speaker oriented occurrences, where Wh-exclamatives mainly serve social purposes and promote some self-staging (2015, 92-93).

Other exclamative markers

8Exclamative illocutionary force is not however restricted to exclamative constructions. It concerns many other linguistic features such as interjections (‘Oh’, ‘Ah’, ‘Ugh’), intensifiers (‘such’ and ‘so’), subject-auxiliary inversion, ‘Isn’t life wonderful!’ (Henry in ‘Something Childish but very Natural’), repetition of nominal phrases, ‘The shortness of life! the shortness of life’ (Jonathan in ‘At the Bay’), exclamation marks. Dashes also frequently co-occur with exclamation in Mansfield:

What can you do if you are thirty and, turning the corner of your own street, you are overcome, suddenly by a feeling of bliss—absolute bliss!—[...] [Bertha] loved Little B so much—her neck as she bent forward, her exquisite toes as they shone transparent in the firelight—that all her feeling of bliss came back again, and again she didn't know how to express it—what to do with it. (Bliss, 116)

9They are indeed many ways exclamation might be communicated. But not all preserve the strength of the enunciative outcry while constructing a linguistic, syntagmatic expression. Some markers thus mimic prosody (italics, exclamation marks), others, like dashes, convey the state of cognitive perplexity induced by intense feeling; expressive markers might be disseminated. Wh-exclamative constructions seem the most accurate linguistic transposition of an exclamation, i.e. the ‘loud, articulate expression’ (in the Oxford English Dictionary’s terms) of a feeling.

Free indirect thought

10How does the Mansfieldian narrator account for such a cognitive disruption without verging on gaudy sentimental literature? Omniscient narration might describe the inner turmoil of the characters, but external focalization often fails to communicate the true perspective, that of the ‘experiential’ centre. This is why the technique of free indirect discourse seems so relevant in the expression (not description) of feelings, as we will see.

11The majority of Mansfield’s short stories are indeed told in the third person. An omniscient narrator might open the story, but his/her viewpoint subtly and frequently gives way to that of one or many intradiegetic characters. Free indirect discourse, or to be more exact in Mansfield’s case, free indirect thought is a frequent mode in her writing, a mode that enables narrator-filtered internal focalisation. Thus, the protagonists’ feelings are better understood than in any direct monologue —where the characters, because they are immersed in emotion, produce linguistic approximations (as we have seen in Phyllis and the narrator of ‘Je ne parle pas français’). Free indirect thought might strike the right balance between emotional intensity (markers of personal perspective) and linguistic construction (omniscient filter). Kuivalainen (2009) underlines that ‘Mansfield often uses clause-initial Wh-elements in exclamations to introduce free indirect discourse in the telling’.

Theoretical framework and method

12Following cognitive grammarians such as Langacker (2009) and Cotte (1998), I believe that grammar is motivated, that grammatical forms are meaningful, and that some specific perceptions/reactions to the extralinguistic world might be understood and retrieved thanks to a contextual study of specific markers. Style arises when such specific markers, in a given textual entity, work as a system and all converge in a unique, specific expression of reality. Wh-exclamatives play an essential role in Mansfield’s style.

13Following (broadly defined) enunciative perspective, psycholinguistics (Guillaume 1973) and embodied cognition (Lakoff 1980, Damasio 2012), I pay close attention to the context of utterance. All speakers are made of flesh and blood and participate in a living, moving environment, which some of their linguistic productions show, especially when it comes to expressing feeling. I also borrow from phenomenological notions such as qualia (the subjective quality of experience or "how it is to experience something" in Konderak's terms, 2016, 292) in my endeavor to account for the subjective dimension of a phenomenon.

  • 3 There are very few occurrences of embedded exclamatives in my corpus (9%).

14In order to define how Wh-exclamatives express feeling, I have gathered all (independent or subordinate3) Wh-exclamative constructions in two collections of Mansfield’s short stories, Bliss and Other Stories (1923) and The Garden Party: and Other Stories (1923). Here is a list of the stories included in the two collections:

Bliss and Other Stories

The Garden Party: and Other Stories


Je ne parle pas français


The Wind Blows



The Man Without a Temperament

Mr Reginald Peacock’s Day

Sun and Moon

Feuille d’Album

A Dill Pickle

The Little Governess


The Escape

At the Bay

The Garden Party

The Daughters of the Late Colonel

Mr. and Mrs. Dove

The Young Girl

Life of Ma Parker

Marriage a la Mode

The Voyage

Miss Brill

Her First Ball

The Singing Lesson

The Stranger

Bank Holiday

An Ideal Family

The Lady's-Maid

15The number of short stories thus studied amounts to 29 (about a third of all Mansfield’s production).

16I thus collected 249 Wh-exclamative utterances. I then analyzed them in-depth from different angles — grammatical, semantic, narrative and cognitive ones —, trying to encompass all relevant aspects of the matter. I first studied the construction of the occurrence per se, then the context of utterance.

17The occurrences were described according to the following criteria:

  • nature of Wh-operator (HOW or WHAT)

  • nature of Wh-phrase (Noun Phrase, Adverbial Phrase or Adjectival Phrase)

  • eventual presence of a modifier before the phrasal head

  • function of Wh-phrase in Wh-structure (Subject Complement, Object Complement, Adverbial function): I found zero occurrence, in my corpus, of a Wh-phrase functioning as Subject in an exclamative construction

  • potential omission of subject and verb

  • nature of subjects when expressed

18The context of utterance was then studied according to the following criteria:

  • punctuation used in Wh-utterances

  • potential presence of interjection just before the utterance

  • potential presence of explicit lexical field of emotion (before or after the utterance, maximal distance with Wh-construction: 2 sentences)

  • speech mode of Wh-utterances

  • type of reaction eliciting Wh-utterances

  • potential subordination of Wh-constructions

19The data collected were then interpreted when relevant. Such a detailed analysis corroborates some well-known facts about Wh-exclamatives such as the frequent omission of subject and verb in the Wh-clause, which, combined to the study of the above-mentioned contextual parameters, might be understood in a new way. My data also enabled me to sketch out some remarkable and yet unstudied differences between HOW and WHAT markers.

20It first appears that Wh-exclamatives mostly express a cognitive reaction to a situation felt with some intensity, which intensity results in indeterminate marker Wh-.

Wh-operator: cognitive intensity and indeterminacy

Wh-marker in exclamative phrases (first statistics and hypotheses)

21The essential, initial component of a Wh-exclamative is the Wh-morpheme, which interrogative and exclamative HOW and WHAT have in common, along with many other interrogative words (where, when, why, who etc). Indeed, in more than a hundred languages, as Haspelmath (1997) has shown, the first historical use of both Wh-operators was an interrogative one. Exclamative uses of Wh-markers are thus traditionally said to derive from interrogative uses while retaining some of its core value: an ‘information gap’ (Lapaire & Rotgé 1994), some lack of knowledge, some indeterminacy.

  • 4 "parcours" in French.
  • 5 Which some scholars relate to their factivecomponent (they cannot be debated nor answered to, (...)

22In interrogative constructions the Wh-marker is often interpreted as representing an open operation of scanning4 (between indeterminate variables), waiting for the answer to ascribe a definite value to the variable. It is indeed intended by the speaker asking a question that such an information gap will be filled. But it is not the case in exclamative constructions which are mainly speaker-oriented, and do not enter into any question/answer pairs5.

23Interrogative and exclamative uses of HOW and WHAT actually differ in other ways. As Huddleston and Pullum (2002) notice, HOW and WHAT are the only two exclamative words available, whereas there is a large number of interrogative words. As such, the two markers play a core role in the expression of the exclamation, a core role in which ‘they are showing some differences from their interrogative counterparts in both grammar and meaning’ (2002, 918). Fronted in all our occurrences (since no exclamative phrase is used as subject in my corpus), Wh-phrases concentrate exclamative vehemence. Indeed, a grammatical difference foregrounded by Huddleston and Pullum is the possibility, for HOW and WHAT, to modify any other degree modifier such as ‘very,’ ‘remarkably’. Doing so, they convey the speaker’s involvement in discourse, in Huddleston and Pullum’s terms (2002, 922), ‘they express the speaker’s strong emotional reaction or attitude to a situation’.

  • 6 The subject of such Wh-exclamatives expressing reaction to a ‘situation’, when mentioned, is ofte (...)
  • 7 I have made a distinction between exclamative reactions towards (an) other character(s)'s behavio (...)

24In my corpus, all Wh-exclamatives involve such a reactive component to a ‘situation’, the situation being either actualized in context, remembered or imagined. I have attempted to describe and classify the contexts of each utterance; 'situation'6 being applied to the description of the context when the speaker’s reaction encompasses more than one single, identifiable element such as a behaviour7 or an idea.

25In 51% of all occurrences, Wh-exclamatives express an affective reaction to such a large situation. The expressive motivation of exclamatives has long ago been emphasized (Searle 1969) and never denied since (Sadock & Zwicky 1985, 162), but it seems to have been reduced in the semantic and pragmatic literature to a feeling of surprise (maybe because of the undeniable reactive component), which, in turn has given rise to a dominant conceptualisation of Wh-exclamations as expressing surprise at ‘violation of expectations’ and being intrinsically scalar (Zanuttini & Portner 2003, Rett 2011, Celle & Lansari 2017).

  • 8 Such a questioning is developed by Nouwen and Chernilovskaya (2015) but mainly for other language (...)

26I would like to question8 these approaches, in the light of my corpus data analysis, and to suggest that:

  1. Wh-exclamatives are not necessarily motivated by surprise

  2. they do not necessarily consist in degree evaluation (their meaning is not necessarily scalar, as Bolinger (1972, 188-189) suggests: ‘Like what, but unlike so and such, how may be used in exclamations that do not intensify but comment on some atypical semantic feature’

  3. they rather express indeterminacy and qualitative identification

How are Wh-expressions expressive? Speaker-oriented interjections

27I define expressivity as the communication of his/her feeling by the speaker, whereby the co-speaker is able to perceive some of the speaker’s emotional involvement. The expression of feeling might be expressive or not; the more it verges on description and full syntactic articulation, the less expressive it generally becomes. I draw here on the somewhat cryptic distinction established by Guillaume between ‘expression' and ‘expressivity', a distinction developed and explained by Legallois and François (2012, 8). Expressivity relies on implicitness and enunciative implication, it has to do with synthetic experiential language, while expression consists in analysis and linearity.

28Only the Wh-marker, in Wh-exclamatives constructions, is expressive. Other elements are less expressive, being more analytical. In the effort to elucidate the feeling he or she has just experienced, the speaker constructs and develops what was at first not rationally construed.

29My corpus data confirm the speaker-oriented dimension of Wh-exclamatives; a study of the narrative mode of utterances shows that most Wh-exclamatives occur in contexts where the speaker either is alone or speaking/thinking to herself/himself.

30I have established, following Mansfield’s texts, 4 categories of discourse mode:

  • narration (N) (external focalization)

  • direct speech (DS) (situation of dialogue)

  • direct thought (DT) (utterance to oneself, close to interior monologue)

  • free indirect though (FIT) (internal focalisation)

31The following diagram shows the repartition of all occurrences according to these criteria:

32FIT is the predominant speech mode (53%) in which exclamative utterances can be found. It confirms Celle and Lansari (2017) characterization of Wh-exclamatives as representing emotional or ‘pathetic expressivity’ which does not concern the co-speaker.

33Miss Brill is lonely and unhappy but every Sunday afternoon, she collects precious fragments of happiness while listening to music, watching people live, thinking they love each other and fantasizing herself as being part of the show:

Oh, how fascinating it was! How she enjoyed it! How she loved sitting here, watching it all! It was like a play. It was exactly like a play. (‘Miss Brill’, The Garden Party, 204)

  • 9 Free indirect discourse allows Modernist writer Katherine Mansfield to render the specificity of (...)

34In free indirect thought9, the narrative voice fuses with that of her character, so much so that it might restitute the expressive vehemence of the feeling they experience, visible in many expressive markers: the interjection ‘Oh', the marker ‘HOW’, the punctuation marks, the triple iteration of the exclamative structure, ending, as it is often the case, in a more explicit (and less expressive) analytical expression of what it is that creates such intense feeling (‘watching it all’).

35Direct thought is another context where the expressive (non dialectical) dimension of Wh-exclamatives is made manifest; overwhelmed by the excitement of her first ball and the beauty of it all, Leila thinks breathlessly, "How heavenly; how simply heavenly!" (‘Her First Ball’). Again, we find a repetition of the Wh-exclamative, this time with the omission of subject-verb, as can be found in such constructions. Repetition of exclamative pattern in a close context (less than 2 sentences) happens in 24% of all occurrences.

36Contextual parameters help us indeed to evaluate expressivity and speaker’s involvement; the study of potential interjections is also interesting, as is that of punctuation.

37It appears that interjections, far from being systematic before Wh-utterances, are not so frequent; similarly, exclamation marks, although predominant, are not always used.

38Expressivity appears in interjections (always just before the Wh-utterances) manifesting the physicality of the speaker’s involvement; but the Wh-marker, in itself, is expressive. 65% of utterances do not present any interjection, and 52% do not use exclamation mark.

  • 10 There might also be a phonetic reason to it, with the realization of the <h> phoneme characterized b</h> (...)

39This can be explained by the core value of the Wh-marker, which continues the interjections when present, and play their role when there are none. The exclamative illocutionary force is transmitted with a Wh-phrase, because the Wh-marker has an interjectional value. Being historically10, semantically associated with question and suspension of determinacy, HOW and WHAT realize a discursive disruption.

40The Wh-marker, though minimally, signals the physical presence of the speaker — but as something else than a speaker, as an experiencer. The fact that there is a frequent omission of verb and subject in Wh-utterances further shows the speaker’s cognitive involvement.

The subjects of exclamatives: a status of synthetic, contextual evidence

Omitted or pronominal subjects

41That Wh-exclamatives might be constructed without their verb and subject is well known. But what this fact reveals about their meaning has not been studied yet. I would like to argue that a completed exclamative structure loses in expressivity what it gains in analysis, that the essence of exclamation rests in the beginning of the structure, in the Wh-phrase, which I will describe, following Cotte’s analysis of the Wh-marker in questions, as both topicalized and focalized.

42Let us first have a look at my data. Not less than 38% of my occurrences are subjectless (and verbless) constructions.

43What do the expressed subjects look like (Fig. 6)?

44Pronouns are used in 77% of all occurrences. Pronouns provide indirect, anaphoric, mediated reference. Their referent must be retrieved in context. The least frequent exclamatives (2%!) are those which take proper nouns as their subject. Referents of proper noun (social, ‘objective’ and full identification) constitute rare subjects of exclamatives. Speakers who exclaim do not typically ‘identify’ what it is that impresses them (a person? a behaviour? a situation? a memory? a mere thought?), because they are very close to the ‘subject’. The ‘subject’, however large its referent may be, is obvious, in its literal sense, it stands in the way of the speaker/perceiver. This is even made manifest by the repartition of the pronominal subjects in our corpus: it is the pronoun which is used the most frequently, before the personal pronouns.

Dominance of IT

  • 11 More analysis on IT might be read in Neveux (2012).

45‘It’ is the pronoun whose reference may vary the most, from a single object to what seems to be a large and vaguely defined situation. Situational ‘it’ has been called ‘ambient it’ by Bolinger (1973), who suggested that ‘it’ in subsuming many contextual parameters might not be deprived of meaning. ‘It’ is understood in context. ‘It’ is the context. What ‘it’ refers to is mostly impossible to define for anyone who is not in context. This is why it is very interesting that ‘it’ (when the subject is indeed expressed) is the most frequent subject of all our exclamative constructions11.

46Miss Brill’s use of ‘it’ in ‘Oh, how fascinating it was! How she enjoyed it!’ indicates such contextual evidence: she feels very well what ‘it’ is about (some feeling of belonging), although she might not know it, as the narrative voice concludes with irony: ‘And Miss Brill's eyes filled with tears and she looked smiling at all the other members of the company. Yes, we understand, we understand, she thought—though what they understood she didn't know’. ("Miss Brill", The Garden Party, 206)

47Intense feelings imply a specific situational involvement that seems evident to the speaker. Exclaiming is articulating such a reaction to an evident situation. This evidence is hard to convey.

48This is why ‘it’ is the adequate pronoun to encompass such a synthetic reference. In ‘Bliss’, Bertha Young suddenly feels overwhelmed by irrepressible pangs of love towards her baby, of whom the Nanny, as usual, is taking care:

The baby looked up at her again, stared, and then smiled so charmingly that Bertha couldn't help crying:

"Oh, Nanny, do let me finish giving her her supper While you put the bath things away.

"Well, M'm, she oughtn't to be changed hands while she's eating," said Nanny, still Whispering. "It unsettles her; it's very likely to upset her."

How absurd it was. Why have a baby if it has to be kept—not in a case like a rare, rare fiddle—but in another woman's arms? ("Bliss", Bliss, 119)

49What ‘it’ refers to in this passage is quite clear: the fact of having someone else to take care of her baby while Bertha would rather take care of her herself. Which is explained in the next sentence, in the form of a rhetorical question. But this very situation entails many concrete parameters, which all trigger Bertha’s anxiety much more than just the idea itself: the baby ‘looking up at her', 'smiling charmingly’, and the condescending opposition of the Nanny which intensifies Bertha’s frustration. ‘It' subsumes, synthesizes all those parameters, in the exclamation, before it is explained in a more analytical, less expressive moment. The after-thought gives articulate form to the complex, indeterminate feeling (desire+frustration).

50The synthetic quality of the pronoun ‘It’ appears when adverbial ‘all’ follows it, as in the following occurrence (the little governess is about to embark on a night train which is to take her to Germany): "How strange it all is," thought the little governess, "and the middle of the night, too." ("The Little Governess", Bliss, 243)

51It therefore seems coherent that Wh-exclamatives might not specify their subject/verb. It is the result of the cognitive (affective) implication of the speaker. The more the speaker is implied, the less analytical the structures will be. Incompleteness is part of the exclamative impulsion, as a transposition of cognitive indeterminacy. Subject and verb might be added, not omitted. Wh-constructions therefore might not be construed as verbless or subjectless. On the contrary, it is the complete clause construction that might be seen as adding verb and subject to the initial exclamation. What matters, for the exclamative illocutionary force to be conveyed, is the expression of the initial Wh-phrase.

52Cotte (in press) argues, following Guillaume’s distinction between resultative syntax and genetic syntax (1973, 216-218), that the Wh-word initial placement in interrogative constructions is the result of a genetic operation leading to the promotion of the Wh-word both as topic (thème in French) and as focus. I agree with such a theory. By fronting the Wh-word, the speaker topicalizes the very lack of information and thus solicits an answer, the topic itself being the indeterminate scanning of variables, such indeterminacy receiving the primary focus of the question. As Cotte himself suggests, the same analysis could be made about Wh-exclamative constructions, where frequent omission of verb and subject further enhances the status of initial Wh-phrase as topic and core focus. Even when subject and verb are mentioned, they are in no way focalized. This is a crucial difference with interrogatives, where subject might receive narrow focus. Compare:

  1. how tall is she?

  2. how tall (she is)!

53In (1), primary focus/nucleus falls on ‘how tall’ and narrow focus/nucleus might fall on ‘she’, while word order in (2) precludes focus/nucleus on ‘she’ and ‘is’. As MacCawley (1998, 713) emphasizes, such a difference between exclamatives and interrogatives is central: ‘stress on the subject is excluded in the so and WH exclamatives’.

54I will now turn to a more in-depth analysis of my corpus data, in order to explain why I question the conceptualization of exclamatives as surprise-based clauses systematically denoting scalar evaluation.

Questioning scalar evaluation: wh-exclamatives express a specific quale

55In my corpus, HOW constructions (161 occurrences out of 249) are much more frequent that WHAT constructions (88/249):

56This asymmetry could be explained by the fact that HOW can be used as modifying an adjectival or an adverbial head in the exclamative phrase, whereas WHAT can only be used with a nominal head. But other elements will be taken into account. Figure 9 shows that the most frequent nature of exclamative phrases in my corpus is the adjectival one, and figure 10 shows that the most frequent syntactic function is the subject-complement one.

57These data might be interpreted (dominance of SC function and Adjectival nature) in a cognitive and phenomenological approach, as confirming the essential meaning of an exclamation: a speaker exclaims to express some reaction to a quality affecting him, a quale in the philosophical meaning, i.e, a quality or property as experienced by the speaker. This quale is perceived with some intensity, which results in the use of indeterminate Wh-marker.

  • 12 Bolinger (1972) thinks differently, as already mentioned. Even if there is no overt degree morpho (...)

58Zanuttini and Portner (2003, 40) propose the concept of "widening" to analyze the specificity of the Wh-operator in exclamative clauses; in the example "What a delicious dinner you’ve made", the authors see 'delicious’ as undergoing a 'domain extension', since the quality thus foregrounded can not be measured according to the available scale of deliciousness: "the speaker implies that the tastiness of the dinner exceeds the range of possibilities previously under consideration", comment Zanuttini and Partner (2003, 54). Many scholars also define exclamatives as expressing surprise at "a noncanonical situation" (Michaelis 2001, 1039, Celle & Lansari 2015, 96), an "expectation violation" (Rett 2011, 440); the specificity of Wh-exclamatives, according to Rett (2011, 412), being this semantic restriction: "they can only receive degree interpretation12".

59Although some exclamatives might express degree evaluation, as in how many/how few exclamatives, I would like to contest this dominant view.

-‘Oh, oh, how mournful, how mournful!’ ("Revelations", Bliss, 269)

60Monica Tyrell’s exclamation in "Revelations" entails many markers of expressive exclamation: double interjection, punctuation mark, iteration of exclamative phrase. But it is not clear what exactly is mournful (omission of subject), and both the obviousness/strength of the quale [mournful] and its indeterminacy are conveyed in the adjectival exclamative. It is grammatically (very) possible to add ‘very’ to the adjectival construction to prove its scalarity; if we take Bolinger’s defining criterium for degree adjectives as accepting lexical intensifiers (Bolinger 1972, 21), without paying attention to the context of utterance, most of our adjectives would indeed be degree ones.

-how very mournful!

61But it seems to me that however grammatical, this change alters the nature of the speaker’s motivation, more apparent if one enlarges the context: Monica has been suffering from ‘her nerves’ all morning and she seeks refuge at her hairdresser’s; but George does not attend to her as he usually does, and some indeterminate feeling (frustration? Disappointment?) inexorably rises in her:

The brush fell on her hair. Oh, oh, how mournful, how mournful! It fell quick and light, it fell like leaves; and then it fell heavy, tugging like the tugging at her heart. "That's enough," she cried, shaking herself free. ("Revelations", Bliss, 269)

62Monica’s feeling is nowhere identified, but the comparison of the falling brush to the leaves and the recurrent mention of the wind blowing outside (before our extract) shape this quale as a melancholy one, as ‘mournful'. The ‘ful’ suffix, although not precluding any scalar interpretation, does not have to be interpreted in such a way. It is already absolute. According to the OED, ‘ful’ when forming adjectives generally means ‘characterized by (what is expressed by the first element)’. Thus ‘mournful’ means characterized by ‘mourn’: the situational quale might be expressed as saliently mournful. Not very mournful, or more mournful than what was expected, for nothing mournful was expected.

63Most contexts of our occurrences do not consist in surprise, although some might entail a surprise component, since intense emotions often come with some cognitive loss of control. Other more prototypical and complex feelings motivate WH-exclamatives in my corpus, such as love, anger, sadness, bliss, excitement, shame, empathy, admiration etc.

64This objection to scalar evaluation in adjectival exclamatives might be even more convincing with the adjective ‘heavenly’ used by Leila in "Her First Ball" (The Garden Party, 208, 211): the young girl finds it very hard not to 'smile too much', not to share her 'rush of joy' and 'excitement' at experiencing her first ball, and, gazing at the red and golden dancing decor, 'breathlessly’ exclaims (in her thought):

- "How heavenly; how simply heavenly!" ("Her First Ball", The Garden Party, 212)

65"Heavenly" is modified, in its second occurrence, by "simply" (not "very"); for the adjective is (almost subconsciously) uttered as referring to a salient characteristic of the ball at that specific moment, the ball being perceived by Leila as simply belonging to the abode of the gods, and "*very heavenly" does not work.

66Other examples of adverbial modifiers following HOW and WHAT in my corpus indicate either high degree or manner:

  1. -"How--very--extraordinary!" said she. ("The Man Without a Temperament", Bliss, 182)

  2. -"How frightfully unreasonable Stanley is sometimes," ("Prelude", Bliss, 17)

  3. -"But oh," she wailed. "What a dreadfully dismal outlook." ("Psychology", Bliss, 151))

  4. -"My word, Laura! You do look stunning," said Laurie. "What an absolutely topping hat!" ("The Garden Party", The Garden Party, 85)

  • 13 Bolinger does present such adverbs as ‘boosters’ (high scale intensifiers) but also notes that th (...)

67'Frightfully’ in (2) and ‘dreadfully’ in (3) are no degree adverbs13 but adverbs of manner specifying further the effect, on the speaker, of the quality ascribed (respectively ‘unreasonable’ and ‘dismal’). They are in fact often described as a subcategory of adjective (+LY), and (2) might then be construed as the expression of a complex quale where the semantic features <frightful> (about the predication) and <unreasonable> (about Stanley) combine.

  • 14 Henkel’s suggestion, in a personal exchange.
  • 15 For the sake of place, adverbial exclamative phases have hardly been mentioned in this study. But (...)

68Scalarity is very difficult to assess. It must be granted that adjectives, being syncategorematic terms, are the most easily gradable part of speech, all the more so when they are used as predicative adjectives after a linking verb, as Bolinger (1972) and Henkel (2015) interestingly note. But the context of utterance must be taken into account in determining scalarity, and other criteria14 might help to establish degrees in scalarity, such as potential antonyms and measurement; for example, ‘big’ can be measured, has ‘small’ for antonym, and there might be a certain degree at which ‘big' might be evaluated as ‘big’ by everyone (there is some objectivity to the assessment of ‘big’). ‘Mournful’ and ‘heavenly’, according to these criteria, are not compatible with any degree evaluation15.

  • 16 In my corpus at least, it seems that surprise is mostly staged in the WHAT exclamatives described (...)

69Most HOW occurrences in my corpus (supposedly most compatible with scalar expression) do not have clear scalar interpretation at all. Most adjectives do not deal with measurement. On the contrary, they reveal a highly subjective, contextual perception of a quale felt as qualitative in essence. Rather than indicating higher degree than expected, they merely express some indeterminate yet salient semantic characteristic. Terrifying, submissive, dreadful, delightful, marvellous, different, tragic, horrible, quiet, typical, curious, wicked, vile, divine, fascinating are some examples of such adjectives in my corpus. Again, grammatical tests might well prove their potential scalarity, but the context almost never shows such an evaluative concern. What the context always does show is a reaction towards a quale, which is perceived as obvious rather than as indicating higher degree of some quality than expected. In my corpus at least, characters never expect some feeling to arise in them. And, again, most utterances are motivated by complex and emotional situations rather than by mere surprise16. Exclamative HOW can be interpreted as manner adverb as much as as degree adverb.

WHAT-exclamatives: social expressivity

70Nouns (as categorematic terms) are not easily gradable. If they are modified, they might become so. In my corpus, as shown in the following NP and AdjP diagrams, most WHAT exclamatives do not have any adjective before the head noun, but they do modify their head more frequently than HOW constructions:

71Two reasons might explain this asymmetry. As said earlier, HOW exclamatives focus on mere expression of a specific (intense yet indeterminate) quale, rendered in the focalized adjective. WHAT exclamatives do not work this way in my corpus. Syntactically, they often consist in a hypernymic head that needs to be modified in order to express judgement. WHAT exclamatives mostly convey judgmental reactions (still not necessarily in reaction to any expectations). Either the head noun already entails semantic (mostly negative in our utterances) judgement and it has no adjective as in (1), or it only works as conceptual ‘shell’ noun as stated by Schmid (2000), hence its hypernymic dimension, and the adjective adds judgmental feature as in (2):

  1. "Oh, What a mistake to have come! Fatal. Fatal." ("Revelations", Bliss, 268)

  2. "Oh, What a loathsome thing to have done." ("Marriage à la mode", The Garden Party, 183)

72Recurrent hypernyms in my corpus are the nouns ‘thing’, ‘thought’, ‘idea’, ‘person’ etc. Evaluative nouns are ‘mistake’, ‘fool’, ‘idiot’, ‘rot’, ‘nonsense’ etc. Such a frequent use of hypernyms is yet another proof of indeterminacy in exclamative constructions, and of the nature of the speaker’s motivation: subsuming such indeterminate traits and expressing them in HOW exclamatives or judging them in WHAT exclamatives. The nominal exclamative constructed with WHAT signals a categorizing motivation; it is the symbolic and specific meaning of the nominal part of speech to provide identifying and clear-cut frontiers to a given entity, as Mignot (2014), following Langacker (2009), demonstrates. "With what, no degree words pose no problem", notices Bolinger (1972, 69). Nominal exclamatives are thus more apt to express social judgement than adjectival ones in HOW, the (unconscious) choice of the nominal part of speech resulting from the same need to categorize and classify situations/behaviours/physical appearances, to discriminate what is proper from what is not. In "Sun and Moon", a cold, unloving socialite mother, who has been sending her two animal-children away ‘squeaking’ all day long in order to prepare her big evening party, finally expresses one single mark of affection (satisfaction) towards them. After they have been transformed into little social freaks with their very heavy party costumes, she exclaims: "What a picture!" ("Sun and Moon", Bliss, 212). 100% of my occurrences expressing reaction at physical appearances are indeed constructed with WHAT marker.

73Other interesting data (listed below) arising from my corpus analysis corroborate Lansari and Celle’s (2015) description of some exclamatives as expressing social ‘proactive’ interaction. It seems that this particular meaning and expressivity, which the authors characterize, following François and Legallois 2012, as ‘ethical expressivity’ (having to do with representation rather than emotion) is mostly realized by WHAT-utterances (in my corpus at least):

  1. the compatibility of WHAT with extraposed subject construction (65% of all extraposed subject utterances, an analytical development which I interpret as showing less emotional expressivity. Some utterances express both subjects, e.g. ‘What a pity it is that you did not arrive by daylight’. ("Je ne parle pas français", Bliss, 104)

    • 17 It is however debatable whether embedded exclamatives can be still considered as exclamatives at (...)

    the incompatibility of WHAT with embedded clauses17 (only one occurence out of a total of 23 embedded exclamatives in my corpus)

  2. the great proportion of WHAT exclamatives in direct speech (47%, whereas they represent only 35% of all occurrences, and while HOW is preferred in all direct thought utterances, where the character thinks to herself/himself)

  3. the greater proportion of elliptic constructions in WHAT exclamatives than in HOW-exclamatives (60% vs only 38% for HOW occurrences)

74WHAT constructions are uttered in context of immediate reaction; they often convey a social/ethical judgment. Because the narrative mode is often direct speech, the speaker and co-speaker ‘share’ the situation, the speaker does not need to explicit the subject and verb. The speaker’s involvement is not as affective as it is in HOW constructions. A study of lexical markers of emotion in the context of utterances indeed shows that 78% of utterances whose context includes lexicalized emotion are constructed with HOW, and only 22% with WHAT.

75HOW exclamatives are mostly devoted to the articulation of emotional introspection.


  • 18 Although Bolinger (1972, 72) interestingly notes that "If the context forces the question to be s (...)

76In the light of my corpus data, I have attempted to question the dominant definition of Wh-exclamatives as always enacting surprise, scalar evaluation and violation of expectation. A hypothesis for the success of such theory might be the assimilation of exclamative to interrogative uses of Wh-marker. In questions, Wh-words can express some quest for measurement18. But exclamatives do not typically evaluate; they express a reaction to some indeterminate yet strong feeling. They might judge and categorize in WHAT constructions, which occur in more social interactive contexts. The frequent omission of verb and subject highlights the core role of the initial Wh-phrase in WH-exclamatives, where expressivity lies.

77WH-exclamatives, especially in free indirect thought, are a major feature of Mansfield’s style: they allow living, intimate expression of minor and otherwise socially silenced characters, they retain some of the cognitive conflict between intense feeling and language (internal focalization) while solving it in the clausal, analytical expansion (external focalization, use of ‘she/he’ instead of ‘I’), they strike a perfect balance between expression and expressivity. They might even embody Mansfield’s interest for our many selves: ‘True to oneself! Which self? Which of my many — well really thats what it looks like coming to — hundreds of selves.’ (Mansfield 1997, 204). Such multiplication/opening of the self is frequently experienced by the writer:

What a QUEER business writing is. I don’t know. I dont believe other people are ever as foolishly excited as I am while im working. How could they be? Writers would have to live in trees. I’ve been this man been this woman. Ive stood for hours on the Auckland Wharf. Ive been out in the stream waiting to be berthed. Ive been a seagull hovering at the stern and a hotel porter whistling through his teeth. It isn’t as though one sits and watches the spectacle. That would be thrilling enough, God knows. But one IS the spectacle for the time. (Mansfield 1977, 199)

78In such 'blazing' moments, such critical and 'central points of significance' (Mansfield's terms, 1987, 89), one might be 'least personal', paradoxically most oneself, and most alive. Exclamatives in free indirect thought enable Mansfield to fulfil her essential artistic ambition: to render the inner, emotional life of her characters without any too explicit representation of sentiment. Her protagonists are not sentimental. No realistic detail ever describes at length their emotional state, which they are unable to exteriorize. But they burn with intense, lively and undefined emotion, to which free indirect exclaimed thought provides direct, unmediated access. Mansfield thus compensates for her characters' emotional isolation by setting free and giving voice to their immense capacity for empathy (feeling with). Her aesthetics is one of the miniature: exclamatives in her short stories give us minimal glimpses at what sensitive, artistic life is about – feeling immensely with.

Haut de page


Primary Sources

MANSFIELD, Katherine. 1923. Bliss: and Other Stories. London: Constable.

MANSFIELD, Katherine. 1923. The Garden Party: and Other Stories. London: Constable.

MANSFIELD, Katherine. 1954. Journal of Katherine Mansfield. J.M. MURRY (ed.). London: Constable & Company Limited.

MANSFIELD, Katherine. 1977. The Letters and Journals of Katherine Mansfield: a Selection. C.K. STEAD (ed). London: Allen Lane.

MANSFIELD, Katherine. 1987. The Critical Writings of Katherine Mansfield. C. HANSON (ed). New York: St. Martin's press.

MANSFIELD, Katherine. 1997. The Katherine Mansfield Notebooks. Vol.2. M. SCOTT (ed.). Canterbury (New Zealand): Lincoln University Press, Wellington (New Zealand): Daphne Brasell.

Secondary Sources

BOLINGER, Dwight L. 1972. Degree Words. The Hague, Paris: Mouton.

BOLINGER, Dwight L. 1973. "Ambient it is meaningful too." Journal of Linguistics 9.2: 261-270. doi:10.1017/S0022226700003789

CELLE, Agnès & LANSARI, Laure (eds). 2017. Expressing and Describing Surprise. Amsterdam: John Benjamins.

CELLE, Agnès & LANSARI, Laure. 2015. “Surprise et exclamation en anglais.” In N. DEPRAZ & C. SERBAN (eds), La surprise à l’épreuve des langues. Paris: Hermann, p. 79-99.

COTTE, Pierre. 1998. L'explication grammaticale de textes anglais. Paris: Presses Universitaires de France.

COTTE, Pierre. In press. « Les questions, wh et la thématisation, une enquête ».

DAMASIO, Antonio. 2012. Self Comes to Mind: Constructing the Conscious Brain. London: Vintage.

DE MATTIA-VIVIÈS, Monique. 2005. "Monologue intérieur et discours rapporté : parcours entre narratologie et linguistique." Bulletin de la Société de stylistique anglaise 25: 9-24. halshs-01021058

GUILLAUME, Gustave. 1973. Leçons de linguistique de Gustave Guillaume 1948-1949. Québec: Les Presses de l’Université Laval, Paris: Librairie C. Klincksieck.

HAMMOND, Meghan Marie. 2016. "Mansfield's psychology of the emotions." In W. MARTIN (author), C. HANSON & G. KIMBER (eds), Katherine Mansfield and Psychology. Edinburgh: Edinburgh University Press, p. 56-67.

HAMMOND, Meghan Marie. 2014. Empathy and the Psychology of Literary Modernism. Edinburgh: Edinburgh University Press.

HASPELMATH, Martin. 1997. Indefinite Pronouns. Oxford Studies in Typology and Linguistic Theory. Oxford: Clarendon Press.

HENKEL, Daniel. 2015. "La syntaxe est-elle déterminée par le sens? Une corrélation statistique entre prédisposition prédicative et gradation comme indice possible." Studia Universitatis Babes-Bolyai Philologia. 3., accessed 3 January 2018.

HUDDLESTON, Rodney & PULLUM, Geoffrey. 2002. The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press.

KONDERAK, Piotr. 2016. "Between language and consciousness: linguistic qualia, awareness, and cognitive models." Studies in Logic, Grammar and Rhetoric, The Journal Of University of Bialystok 48.1: 285-302., accessed September 2017.

KUIVALAINEN, Paivi. 2009. "Emotions in narrative: a linguistic study of Katherine Mansfield." The Electronic Journal of the Department of English at the University of Helsinki 5.9,’s-short-fiction-paivi-kuivalainen/, accessed 2 January 2018.

JAMES, William. 1990 [1890]. The Principles of Psychology. Bristol: Thoemmes; Tokyo: Maruzen.

LAKOFF, George & JOHNSON, Mark. 1980. Metaphors We Live By. 2nd edition. Chicago: University of Chicago Press.

LANGACKER, Ronald W. 2009. Investigations in Cognitive Grammar. Berlin: Mouton de Gruyters.

LAPAIRE, Jean-Rémi & ROTGÉ, Wilfrid. 1994. Séminaire pratique de linguistique anglaise. Toulouse: Presses universitaires du Mirail.

LEGALLOIS, Dominique & FRANÇOIS, Jacques. 2012. “Définition et illustration de la notion d’expressivité en linguistique.” In C. GUIMIER, N. LE QUERLER, F. NEVEU & E. ROUSSEL (eds), Relations, Connexions, Dépendances: hommage au professeur Claude Guimier. Rennes: Presses universitaires de Rennes, p. 197-221.

MCCAWLEY, James D. 1998. The Syntactic Phenomena of English. 2nd edition. Chicago: University of Chicago Press.

MICHAELIS, Laura. 2001. "Exclamative Constructions." In M. HASPELMATH, E. KÖNIG, W. ÖSTERREICHER & W. RAIBLE (eds), Language Typology and Language Universals: An International Handbook. Berlin, New York: Mouton de Gruyter, p. 1038–1050.

MIGNOT, Elise. 2014. "Le sens des lexèmes adjectivaux dans les noms construits de l'anglais: une réflexion sur le nom en tant que partie du discours." In R. DAVAL, P. FRATH, E. HILGERT & S. PALMA (eds), Res per Nomen IV, Les théories du sens et de la référence - Hommage à Georges Kleiber. Reims: Epure Editions et Presses Universitaires de Reims, p. 237-252.

MOURANT, Chris. 2017. "Modernist emotions: the critical writings of Katherine Mansfield and Virginia Woolf.". In T. MARTIN (ed.), Katherine Mansfield and the Bloomsbury group. London: Bloomsbury, p. 161-80.

NEVEUX, Julie. 2012. “It girl, it bag et le pronom it: le sens du pronom neutre non anaphorique.Anglophonia/Sigma 32: 63-84.

NEVEUX, Julie. 2013. John Donne, le sentiment dans la langue. Paris: Rue d’Ulm.

NOUWEN, Rick, & CHERNILOVSKAYA, Anna. 2015. “Two types of exclamatives.” Linguistic Variation 15.2: 201-224.

RETT, Jessica. 2011. "Exclamatives, degrees and speech acts." Linguistics and Philosophy 34.5: 411-442.

SEARLE, John. 1969. Speech acts: An Essay in the Philosophy of Language. Cambridge: Cambridge University Press. doi: 10.1017/CBO9781139173438

SADOCK, Jerrold & ZWICKY, Arnold. 1985. “Speech act distinctions in syntax.” In T. SHOPEN (ed.), Language Typology and Syntactic Description. Cambridge: Cambridge University Press. Vol.1, p. 155-196.

SCHMID, Hans-Jörg. 2000. English Abstract Nouns as Conceptual Shells: From Corpus to Cognition. Topics in English Linguistics 34. Berlin, New York: Mouton de Gruyter.

ZANUTTINI, Raffaella, & PORTNER, Paul. 2003. "Exclamative clauses: At the syntax-semantics interface." Language 79.1: 39-81.

Haut de page


1 Onomatopoeia, often used by Mansfield, are described as “mimetic expressivity” by Legallois and François 2012.

2 Celle and Lansari (2017, 241) show that surprise is not a prototypical emotion in that "it entails an epistemic judgement".

3 There are very few occurrences of embedded exclamatives in my corpus (9%).

4 "parcours" in French.

5 Which some scholars relate to their factivecomponent (they cannot be debated nor answered to, and can only be embedded under factive verbs such as realize’).

6 The subject of such Wh-exclamatives expressing reaction to a ‘situation’, when mentioned, is often ‘it’ or a subordinate clause with propositional content.

7 I have made a distinction between exclamative reactions towards (an) other character(s)'s behaviour and those towards the speaker's (S' in Figure 1) own behaviour.

8 Such a questioning is developed by Nouwen and Chernilovskaya (2015) but mainly for other languages than English.

9 Free indirect discourse allows Modernist writer Katherine Mansfield to render the specificity of the stream of consciousnessexperienced by a character; the term was coined by William James (1890) who insisted on the fluidity of such a cognitive process, thoughts, be they conscious or subconscious, succeeding to sensations, ideas or emotions in a uninterrupted movement. De Mattia-Viviès (2005, 11) specifically characterizes the stream literary technique as reproducing an often elliptic, internal, monologic discourse, un-addressed and un-constructed.

10 There might also be a phonetic reason to it, with the realization of the <h> phoneme characterized by phoneticians as a voiceless glottal fricative that phonologically resembles a consonant but lacks its usual phonetic characteristics. The aspiration thus produced does create a brief suspension in the consonantal chain.

11 More analysis on IT might be read in Neveux (2012).

12 Bolinger (1972) thinks differently, as already mentioned. Even if there is no overt degree morphology in the exclamative, Rett (2011) imagines complex contextual scenarii where covert gradable properties might be added.

13 Bolinger does present such adverbs as ‘boosters’ (high scale intensifiers) but also notes that they are ‘scarcely grammaticized’ (1972, 18); he later emphasizes the importance of context to determine whether adverbs are used as degree or manner adverbs.

14 Henkel’s suggestion, in a personal exchange.

15 For the sake of place, adverbial exclamative phases have hardly been mentioned in this study. But identification of a qualitative (and indefinite) manner (of predicates) seems at stake in our occurrences rather than high degree evaluations, as in the two following examples: "and how his little blue eyes gleamed behind the spectacles!"(where the young girl is overwhelmed with confused and perplexed horror as her German benefactor turns into a lustful monster, "The Little Governess", Bliss, 259); "And wasn't he teaching them all to escape from life? How he sang!", in which singing teacher Reginald Peacock, while his too prosaic wife is snoring at his bedside, proudly remembers and relives his very triumphant day ("Mr Reginald Peacocks Day", Bliss, 206).

16 In my corpus at least, it seems that surprise is mostly staged in the WHAT exclamatives described below as cases of social interaction.

17 It is however debatable whether embedded exclamatives can be still considered as exclamatives at all, given the disappearance of exclamative force in such cases.

18 Although Bolinger (1972, 72) interestingly notes that "If the context forces the question to be specifically degree, what is unacceptable".

Haut de page

Pour citer cet article

Référence papier

Julie Neveux, « Grammar and feelings: a study of Wh-exclamatives in Katherine Mansfield’s short stories »Études de stylistique anglaise, 12 | 2018, 193-222.

Référence électronique

Julie Neveux, « Grammar and feelings: a study of Wh-exclamatives in Katherine Mansfield’s short stories »Études de stylistique anglaise [En ligne], 12 | 2018, mis en ligne le 19 février 2019, consulté le 14 août 2022. URL : ; DOI :

Haut de page


Julie Neveux

Université Paris IV Sorbonne


EA 7332

Julie Neveux is a lecturer in English linguistics and literature at Paris-Sorbonne University. She is a member of the research center CeLiSo (Centre de Linguistique en Sorbonne). Initially trained in philosophy at the ENS Ulm, she develops a phenomenological approach of language, semantics and style, defining style as a phenomenon whose effects rely on the perceptibility, in discourse, of the speaker’s motivation (embodied cognition). Her research interests include psycholinguistic and cognitive analysis of the relation between language and feelings.

Julie Neveux est Maître de Conférences en linguistique et littérature anglaise à Paris IV Sorbonne. Elle est membre du centre de recherche CeLiSo (Centre de Linguistique en Sorbonne). D’abord formée en philosophie à l’ENS d’Ulm, elle développe une approche phénoménologique du langage, de la sémantique et du style. Elle définit le style comme un phénomène dont les effets reposent sur la perceptibilité, en discours, de la motivation du sujet parlant (cognition incarnée). Elle travaille notamment à l’étude psycholinguistique et cognitive de la relation entre langage et sentiment.

Articles du même auteur

Haut de page

Droits d’auteur

Tous droits réservés

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search