How do children learn to use language to express emotions? As most human behavior, language is context-dependent. When studying language in relation to emotions and their nonverbal expressions this becomes exceedingly clear. Without a contextual grounding there is no way to tell whether a swearword is expressing an emotion, if a cry reveals sorrow or fear, if a smile is one of affection, mockery or a sign of bowel movements (e.g. Wolff 1987; Harrigan et al. 2005). Earlier research in verbal and nonverbal emotive expressions has, mainly through thorough context descriptions, shown that young children, like adults, express and interpret emotions through prosodic traits (Weeks 1971; Papousek et al. 1990; Fernald 1993; Goodwin et al. 2012); that children begin to add words to their nonverbal emotive acts early on (Miller and Sperry 1987; Bloom 1993; Gerholm 2007) and that the sociocultural environment of the child is important for what conventions of emotive expressions the child will encounter (Clancy 1986; Schieffelin 1986; Howell 1989). Still, we are far from a scientific understanding of the processes involved and how the relation between nonverbal, vocal and verbal means of expression can be understood, regardless of cultural differences of vocabulary. In the area of verbal emotive expressions, we are also far from having identified the various means available. Is the semantics the key (e.g. Wierzbicka 1993)? Are specific prosodic traits essential (e.g. Fernald 1993)? Or do the contextual clues of a specific situation hold the answer (e.g. Sandlund 2004)?
The aim of this paper is to highlight and describe the forms of verbal emotive utterances that appeared in a longitudinal corpus of 11 Swedish children interacting with parents, siblings and friends. By classifying and labeling verbal expressions as emotive in different ways, it is shown how we can gain a better understanding of emotive language used, but also that we access a way to compare and investigate emotive language in a more thorough manner. The questions addressed are:
What is the relationship between children’s nonverbal/vocal emotive expressions and language?
What kind of verbal emotive expressions do children use as they interact in home environment with parents and siblings?
How do we go about studying verbal emotive expressions?
The article has the following structure: Section 2, Background, describes prior research on verbal and nonverbal emotive expressions, linguistic conventions and cognitive and socio-emotional maturation in young children. Section 3, Method, divided into two subsections, includes information on participants, recording conditions, transcription and analyzing procedures. Section 4, Results, is divided into three parts. The first two parts introduce and describe the forms of emotive expressions that make up the major finding of the study: Descriptive utterances (4.1) and Accompanying utterances (4.2), whereas the third part, Development (4.3), addresses the developmental path these utterances are argued to follow. Section 5, Discussion, returns to the research questions posed and addresses them from a developmental perspective. The article ends with a short summary in Section 6, Conclusions.
There are at least three areas we have to keep in mind and include when trying to describe and categorize a developmental path for verbal emotive expressions. These are Emotions as such; Conventions and language; and, Developmental aspects relating to cognitive and socio-emotional maturation. In the following sections these areas will be described sequentially.
Emotions and their expressive forms are built-in devices of humans’ expressiveness (Damasio 1999; Panksepp 2000; Mascolo 2009). Emotions “as such,” or as “states,” can be defined as particular changes in the constant flow of somatic and/or neurophysiologic activity. The result of a change will cause hormonal and neuronal reactions that in turn might affect body posture, facial movements, cognitive activity and vocal behavior (e.g. Lewis 2000). These reactions could be triggered by inner as well as outer stimuli and lead to specific reactions in our bodies. Some of these reactions can be experienced as feelings whereas others probably pass by without us reflecting on them. Some of these reactions will further be visible to others, for example blushing, turning pale, trembling. Others, that in themselves are not visible, will have consequences that become apparent to others as well. This is the case with the prosodic changes occurring as a consequence of the smooth musculature being affected (Bull 2002). Some prosodic clues inform us that speakers are in an emotional state regardless if they topicalize these emotions or not. There are individual differences in how often and intense we get emotional in a way that affects our vocal cords, as there are also differences in our ability to control these somatic reactions (Denham 1998).
Many nonverbal and vocal expressions of emotions have been identified and described, such as laughter (Wolff 1987; Masataka 2003), crying patterns (Golub and Corwin 1985; Wolff 1987), and facial displays of emotion (Ekman and Friesen 1978; Oster et al. 1992). Through nonverbal and vocal expressions of emotions we get information about subjects’ emotions on-line, and although nonverbal and vocal expressions of emotion can be faked, or misleading, they appear to be hard to eradicate completely. If there is an emotional reaction, in most cases it will show (Hess 1965, 1975; Fridlund 1997; Bull 2002). However, in order for us to interpret an expression as a specific emotive expression (positive/negative; strong/weak; etc.) we need contextual clues (Planalp 1998; Cauldwell 2000).
Turning to language, there are word forms that, more than others, are connected to our interpretation of emotions. These are swearwords, such as expletives and invectives, diminutives, reduplifications and more or less fixed and established constructions like metaphors (e.g. Kövecses 2000). Unlike the physiological emotive reactions, labels and word forms will not automatically inform us whether there is an emotion being expressed as we go through empirical data. Nor will they aid us in the search for other words and phrases used emotively. Even more than for nonverbal emotive expressions, contextual clues are indispensable for our intuitive judgments of emotive utterances. Another difference between verbal and nonverbal emotive expressions is that nonverbal emotive expressions have been studied for a long time, whereas descriptions and studies of the verbal emotive expressions are more recent (Sandlund 2004; Cekaite 2012; Goodwin et al. 2012). Mutual definitions and working procedures are hard to establish when the subject matter is diverse and unclear; however, there have been attempts to agree on a common ground in research on both verbal and nonverbal emotive expressions (e.g. Ochs and Schieffelin 1989; Caffi and Janney 1994; Harrigan et al. 2005) and the present study is another attempt.
2.2 Conventions and language
A possible explanation for the difference between verbal emotive utterances and nonverbal/vocal expressions is that language in and by itself constitutes a socially adjusted code. Alongside learning the language of our community, we learn when and how we are expected to say what (Schieffelin 1979; Ochs 1986; Gerholm 2007). Parents have been shown to go to great lengths in establishing the norms of social verbal behavior in their children (Clancy 1986). There are fixed phrases to apply in specific situations such as condolences for grief, congratulations for others’ success and happiness, etc. During childhood children are introduced to, and witness the use of, these phrasings and learn to use them correctly in situations of their own. However, according to prior research, the implementation of these social norms is among the hardest to grasp for the child (Kurian 1986; Shaw 2004).
Parental training of bodily and vocal manifestations exists as well (such as “don’t chew with your mouth open”, “don’t pinch your sister”, “boys don’t cry”, “don’t point at people”, etc.), but little is known about how far this training goes and if children worldwide start out with the same set of nonverbal acts (such as throwing themselves on the ground when angry) or if bodily emotive behavior is learned in the same manner as language – i.e. by listening to/watching others and adapting the already present repertoire of human language vocalization in terms of phonemes (e.g. Dehaene-Lamberz and Gliga 2004) or body motions in the case of nonverbal behavior.
Socialization practices go hand in hand with the child’s cognitive and socio-emotional maturation, areas and capacities in turn aided and guided by language acquisition (e.g. Gleason and Weintraub 1976; Clancy 1986; Ochs 1986; Shaw 2004). The fact that adjusting to societal norms constitutes a process implies that we, by studying young children’s emotive expressions, have an entrance to the verbal means of displaying emotions before these means have merged completely with the socialization forces constantly active – if this is what they actually do.
Cognitive growth facilitates and is facilitated by language acquisition (Vygotsky 1978; Shields 1979; Jacques and Zelazo 2005; Gauvin and Perez 2006; Kuhn et al. 2014). Concurrently, as one learns the phrases and behaviors appropriate in society, one learns to master impulses related to emotions. However, as the physiological reactions will still be there, they will be noticed due to prosodic changes or bodily appearances even if they are not formulated and expressed in words. There are many studies on the pace with which children get control of their actions and conduct, and it is well known that the development is dependent on the interactions taking place with the people surrounding the child (e.g. Schieffelin 1979; Thelen and Smith 1994). No language, and not much socio-emotional maturation, is possible without the scaffolding frames offered and maintained by parents, siblings and other humans. During this development we would expect to find different levels of competence in, for example, the realm of cognitive maturation, in language, and in socio-emotional maturation. But we would also expect there to be similarities and some typical pace with which children acquire different skills. By studying children on different levels, or in different phases, of development we can get a picture of how the general process looks (Vygotsky 1978; Thelen and Smith 1994).
Summing up, based on earlier research it is presupposed that nonverbal and vocal manifestations of emotions function as our visual and audible guide to when an individual expresses an emotion. This appears to be so whether the verbalizations (potentially) used describe the emotion or the emotive event or not. We do not need a child to utter “I am in pain” in order to interpret a fall followed by a cry as an emotive expression. But we do need context in some form to decide what meaning an expression or utterance has (Sacks et al. 1974; Psathas 1995; Linell 2009). Through prior research we know that humans, in general, are very apt in the identification and interpretation of others’ nonverbal expressions of emotions as these come through in bodily, facial or vocal traits (e.g. Hess 1965, 1975; Fernald 1993; Fridlund 1997; Bull 2002). This ability to interpret nonverbal expressions appears to be affected only to a limited extent by cultural differences (Ekman 1989; Juslin and Laukka 2003; Thompson and Balkwill 2006). We do not know much about how humans interpret and understand verbal emotive expressions, or exactly what shape these expressions take, other than that most languages are likely to have words for basic emotions (like anger, love, joy, etc.), and that there are different words and morphosyntactic ways in which languages can add emotive content to an utterance (Ochs and Schieffelin 1989).
From prior research we also know that children mature in the area of emotive expressions and that the societal norms for how to express different kinds of information are internalized gradually. By studying the actual expressions used as children interact with parents and siblings, we have an opportunity to describe and understand the developmental path of children being socialized into a cultural context.
This section is divided into two subsections: description of the participants and recording procedure (3.1 Data), followed by the analyses made (3.2 Analyses).
The material consists of video recordings of five sibling groups together with one or both of their parents. The recordings were conducted in the children’s homes at regular intervals during approximately a period of two years. The 11 children were in the age span 0;9 to 5;10. The participating families were enrolled through the researcher’s personal connections. The families received copies of all video recordings with their children, but had no further profit from participating. Parents were informed of their right to withdraw from the study at any time. All names have been changed and transcripts and video recordings are handled and kept in accordance with the regulations from the Ethical Review Board, Karolinska Institutet, Stockholm.
Table 1 illustrates the participating children and their ages. The thick borders indicate the constellations in which the children were recorded. In all cases but the last, it also shows the family relationship. The last group consists of two sibling pairs recorded together, with parents from both families present. All families lived in or close to Stockholm, Sweden. Of the ten parents, seven have a university degree and the remaining three have high school level of education. No child had any indication of delayed development at the time of recording, and later inquiries have shown them to grow up as healthy individuals. All children had Swedish as their first language and two of the children had Greek as a second language. Swedish was used throughout the recordings.
All recordings were conducted by the researcher. During the recordings the researcher followed the children around, although occasional attempts were made to keep them indoors. The researcher stayed out of direct interaction as much as possible but intervened in more severe conflict situations if no other adult was present. On average 45 minutes of recording were collected each session. The video-recorded material was transcribed by the researcher and the transcriptions include verbal and vocal transcriptions and annotation for gesture, gaze and activity/context (including descriptions of facial expression or manner of movement).
The working procedure in the current study was to classify an utterance as emotive based on the following traits:
An utterance contained a conventional emotive expression (e.g. “I was sad”; “he loves to jump”; emotive metaphors).
An utterance contained a less conventional but semantically emotive-related expression (e.g. “yuckifish”, “poohead”).
An utterance was delivered together with nonverbal emotive expressions like laughter, crying, screaming, sobbing or together with a prosodic contour of a similar emotive character.
All instances that were regarded emotive (based on the conditions above) were analyzed further in terms of verbal content, context and the child’s age. Expressions that were only nonverbal are not considered further here.
All recording sessions contained emotive expressions. The majority of utterances were regarded as emotive based on the utterance being said while crying, laughing, or because the prosody and contextual information was highly emotive. These utterances were called Accompanying utterances, as the emotive content was heavy even if the semantics of the words was not necessarily emotive. A small number of utterances were regarded as emotive based on the semantics of the words and phrases used. These could have emotive prosody or be combined with tears, laughter, etc., but did not need to in order to qualify as emotive expressions. These utterances were labeled Descriptive utterances.
The children were divided into age groups in order to facilitate developmental comparisons. The grouping was based on closeness in age, but I also strived toward having an equal number of children in each group. A consequence is that while some groups have an age span of one year, other groups’ age span is only five months. Table 2 illustrates the age groups and how the two categories of emotive expressions occurred in the data. It also indicates in what way the utterances were regarded emotive, through conventional semantics or by prosodic traits.
There are very few Descriptive utterances in the data. Among these, the most common form is an older sibling describing a younger sibling’s behavior by adding an emotive label, e.g. “he’s angry” or “he loves to (activity)”. On the other hand, the vast number of Accompanying utterances in the data did not include emotive labels at all. Some utterances were conventional in that they included words like “stupid”, “silly”, “ugly”, “pretty” – words that can be used to describe others in condescending or appreciative ways – but when they appeared emotive it was due to the prosodic traits rather than the words used. This was concluded as there were also examples of “stupid”, etc., being used in contexts where there were no emotive displays in any other form, thus making the words non-emotive as well. A possible explanation is that although context is always needed in order to interpret emotive content, some words are more dependent on context than others (cf. Wierzbicka 1993). Constructions like “I am happy” are emotive regardless of context. However, this does not tell us if the speaker experienced any particular emotion, as it could be uttered ironically, as reported speech or while reading aloud. Adjectives not directly describing an emotive state in someone are, on the other hand, utterly dependent on context to get an emotive interpretation. This context dependence is also shared by the rest of the vocabulary where every single morpheme could have an emotive touch if uttered in an emotive manner. In the present study this difference in context dependence was regarded as an additional tense, calling it Emotive Present when prosodic cues and/or nonverbal emotive expressions led to the emotive interpretation.
Below I provide examples and further description of the two groups and their relation to children’s ages, followed by a commentary on development (4.2.3).
4.1 Descriptive utterances
“Descriptive utterances” is a label used previously for verbal emotive expressions in general (e.g. Wittgenstein 1953; Anward 1986). Descriptive utterances consist of expressions where terms of emotions are named either directly (e.g. “angry”, “sad”, etc.) or by the use of an established metaphor (“to explode”, “to have a break down”, “burst with laughter”, etc.) and were considered emotive due to semantic conventions. Descriptive utterances are produced either while still in an emotive state of some sort, or as explanations and descriptions relating to emotive events and experiences of the past or future. In the data there was one early occurrence of a Descriptive utterance from a child aged 2;1; however, the utterance is a verbatim repetition of a toy uttering “I love you” as you pull a string. In addition, the girl was prompted to the utterance by her father saying “what would you say in reply?”, to which she answered “I love you too”. Following that occurrence, the next age at which a Descriptive utterance appears is 3;2 (illustrated below. For a transcription key, see the Appendix).
Ylva at age 3;2 has had a quarrel with her older sister and is now left with the researcher in the living room as her sister and mother have left for the bathroom. Both girls have been very upset and have cried extensively. Ylva has now stopped crying but is still panting and sobbing while she tries to locate a paper to draw on.
Ylva (Y, 3;2) and the researcher (T) in the living room.
|1||[Astrid and the mother have gone to the bathroom, Ylva is fiddling with some papers on the living-room floor][Ylva is sobbing after her cry]|
|2||Y:||seru att ja e lessen nu?||you see that I’m sad now?||[Said to the researcher T]|
|3||T:||ja: vaför eru lessen?||yeah why are you sad?|
|4||Y:||däför xxx astrid rita på min teckning.||coz xxx Astrid drew on my picture.||[She walks into the kitchen after the others]|
Ylva herself brings the emotion to the surface and names it. On request, she also gives an explanation to the causes behind it. Her understanding of the connection between expression, tears, and emotions also surfaces verbally through her question seru att ja e lessen nu? (‘You see that I’m sad now’). She describes a current emotion although the peak of the emotive event has passed. She is well aware that I have witnessed the scene and the crying.
The second example is also an instance of a child naming a recent emotional experience. In this case, however, the process seems to be more than just informative in that Eva (3;5) labors on the experience and the feeling it has evoked. Prior to this sequence, Eva has been pushed into a door by Richard (4;6). The event is managed on the spot and both children’s mothers are involved. Nevertheless, Eva keeps returning to the subject, now turning to her older brother Klas, who was present when the action took place. There is period of a few minutes when Eva leaves the room, tending to other things, then returning (indicated by - - in column one).
Eva (E, 3;5), her brother Klas (K, 5;7), Petra (P, 2;8) and Richard (R, 4;6) in Eva and Klas’s room.
|1||E:||rakt i dörren å så slog ja mej i huvet (.) visst ja va rädd då? visst va ja rädd då klas?||right in the door and then I hurt my head (.) I was scared then right? I was scared then right Klas?||[She sits by the railroad and tells Klas about what has happened]|
|2||K:||de va faktist JÄTTEDUMT (.) å lägg förresten inte den där här.||that was really STUPID actually (.) and by the way don’t put that here.||[Said to Richard who is standing next to him holding a house]|
|3||E:||visst va ja rädd?||I was scared alright?||[She goes and sits by the railroad again]|
|4||E:||visst va ja rädd när richard slog mej i huv. rakt i dörren, då va ja rädd (..) då blev ja rädd.||I was scared alright, when Richard hit me in the head. right in the door, I was scared then (..) I got scared.||[She rubs her head] [She points to the door, mainly addressing Klas, but the boys are ignoring her]|
Focusing on the verbal utterances, it is noteworthy that Eva uses the term rädd (‘scared’) when describing her experience of hitting the door. Furthermore, she frames it as a question as if she is not entirely sure of what she felt and needs some confirmation.
The linguistic level of these verbalizations places some demands on the speaker as s/he needs a vocabulary including specific emotion labels; s/he needs to be able to identify a bodily experience and map it to that particular linguistic label; and s/he needs to be detached enough from the activities causing the emotions so that s/he can use that knowledge in an interaction. These linguistic, cognitive and socio-emotional demands might be a reason for descriptive utterances being few in the material and metaphorical emotive expressions being nonexistent. When interviewing some of the older children (4;6 to 5;10 years old) quite informally around topics such as “how did you feel when…”, these frame-related expressions could be elicited, but in the on-line daily interaction in the families these were not the expressive forms appearing when the children were in one emotive fit or another. As they did not appear between emotive fits either, an interpretation is that descriptive expressions of emotions are infrequent in children’s everyday language use. Even in the interviews, the children tended to show rather than utter the emotion except for “physiological needs” like hunger, thirst, tiredness, etc. which were labeled and described verbally. Thus, when asked “how did you feel when you received X?” (X being a longed for toy), the child would grin and adopt an exaggerated smile rather than answer “I felt/was happy”.
4.2 Accompanying utterances
Vocal and/or nonverbal emotive expressions co-occurring with verbalizations were of different kinds and used by all children and in all ages studied. All Accompanying utterances were used in Emotive Present, that is, the prosodic traits/nonverbal context makes them appear emotive. However, there was a difference in the kinds of words used, also relating to the ages of the children using these words. This is why the section has been divided into two subsections: Accompanying utterances where the words used were of all kinds (4.2.1 Emotive prosody and context); and Accompanying utterances that had more conventional emotive semantics (4.2.2 Emotive conventions).
4.2.1 Emotive prosody and context
Even the youngest children use vocal means to express emotions, needs and wishes. The most intense of these that appeared in the data were labeled shrieks, i.e. shrill vocal calls often produced in joy or anger. However, as children get older, they start adding discernible words to these calls. Sometimes it is sounds more than words but we also find words like “look!”, etc. (i.e. imperatives in form but not always directed to someone). Below we see Richard producing a series of shrieks while jumping with excitement having discovered a toy, only then going verbal.
Richard (R, 2;1), Klas (K, 3;2), Eva (1;0) and the researcher in Klas’s playroom.
|1||R:||IHI::: (.) IHI:::||IHI::: (.) IHI:::||[Richard produces shrieks of excitement/joy]|
|2||R:||TITTA EVA||EVA LOOK|
|3||[Klas comes and sits down by the toy and starts “showing” Richard how it is done. Richard backs up]|
|4||R:||PLING PLING||PLING PLING|
Richard’s vocalizations start out with only vocal sounds of joy. Seconds later, with the same high pitch, he utters “Eva look” and “pling pling”. The difference is that while the two latter utterances consist of language, the first vocalization is regarded as nonverbal-vocal only. In traditional linguistic descriptions, there is a marked difference between vocalizing in general (like most mammals and birds) and in language (possessed by humans only). In an ontogenetic developmental perspective, however, there is no huge step involved in Richard’s choice of a shriek at one occasion and a “look” at another. The utterances and sounds fill the same function and work toward the same end, whether this is informing others of the nice toy discovery or the child simply being overwhelmed by an emotional experience.
The actual words used in situations like the one above may not be important. What makes the utterance emotive is the prosodic contour coupled with the contextual clues. When undertaking strict phonetic measurements of the shrieks in the material, as well as the one-word utterances sometimes taking their place, we might find differences relating to them being of a positive or negative character (cf. Damasio 1999; Nelson 2005; Wilson et al. 2005). However, in a naturalistic study like the one presented here, it is hard to tell what in the situation of rich contextual information informs us that the sequence is joyful.
Among the first words children in the study uttered emotively were “mommy”, “mine”, “no/not/nothing”. Taking “mommy” as an example, it was used in three different ways. One was a referential and directive use (e.g. “there’s mommy” or a pointing finger and “mommy”), which was not regarded as emotive in the typical case. The other was as an attention-getting device, and in this case, it was often emotive (e.g. “Mommy! Come!”). The third use was as an invocation in which the emotive content was heavy and a prerequisite, as illustrated below.
Richard has had an extended argument with his older friend, Klas. They both want the same tractor and Richard being the younger one has problems to protect his interests. As the sequence starts, Richard’s mother (Å) is trying to divert Richard’s attention and make him focus on something else. Klas’s mother (C) is sitting next to them.
Richard (R, 2;1), his mother (Å) and Klas’s mother (C) in the kitchen.
|1||Å:||den hära (.) e det den stora gräsklipparn?||this one here (.) is that the big lawn mower?|
|2||[Richard picks up a big tractor and throws it]|
|4||C:||vet du va, kom richard ska jag visa dej en sak.||you know what, Richard come here and I’ll show you something.|
|449||R:||mamma papa||mommy daddy||[Richard follows Cecilia, crying]|
Richard’s use of “mommy daddy” is uttered while he slouches away after another mother. His own mother is right next to them, but he does not make any moves in order to resort to her; on the contrary, he walks away from her.
When the child gets older, the one- and two-word utterances are no longer the only option to express emotions verbally; but still, the most frequent expressions in the data are emotive due to prosodic clues and contextual information alone. An example is Samuel at the age of 4;9. In the illustration below, he gets sad and expresses this through nonverbal displays in the form of body posture, which is misinterpreted by the researcher. When this misunderstanding becomes apparent, Samuel switches to verbal expressions and describes how the experienced emotion came about, albeit never mentioning the emotion as such.
Samuel, his two sisters and two neighboring boys have been choosing sides for a soccer game. When this is done – ending in Samuel being in the team with his sisters – the other children set off to start the game. Samuel remains with the researcher, sinking down on his knees.
Samuel (S, 4;9) together with the researcher (T) in the garden.
|1||[Samuel sinks down on his knees in the grass; looking down]|
|2||T:||va ere? gjorde du illa dej?||what’s up? did you hurt yourself?|
|3||S:||näe dom där killarna lämnade mej||noo, those guys they left me|
|4||T:||lämna dom dej?||they left you?|
|5||S:||jaa (..) ja vill (.) då e nån annan på dens lag, då xxx å då sa han till mej du e på mitt lag.||yeah (..) I wanna (.) then someone else is on it’s team, then xxx and then he said to me that you’re on my team.|
|6||T:||mm vill du va på samma lag som pojkarna?||uhm you wanna be on the same team as the boys?|
|7||S:||mm men dom bara säjer att att att han e på mitt lag, och dom andra e på mitt lag, så bara säjer men ingen säjer att men ingen vill va me på mitt lag.||mm but they just say that that he’s on my team, and the others are on my team, they just say like that but nobody says that, but nobody wants to be on my team.||[affected voice][affected voice]|
Summing up the type of verbal emotive expressions where the main emotive part was prosodic and context-tied, we have:
one-word onomatopoetic sounds, exclamations or words with a clear emotive prosody (“ihi:”, “look”, “mommy”, “nothing”);
longer utterances where the context and the prosodic traits expressed emotions (“they left me”).
The function of the first kind was similar to interjections but with innovative words/sounds. In a few cases, the function instead resembled invocations but with significant others in the place of Lord/God. The function of the second kind of prosodically emotive Accompanying utterances was much more diverse. In the example given here (ex. ), Samuel was elaborating feelings experienced, prompted by a question from the researcher. Other examples in the material were “I promise not to” (crying reply to a verbal threat from a parent), “I’m gonna hit you” (one angry child to another in the midst of playing), “all the monkeys in the bed” (one child to another while laughing and playing). Accompanying prosodic utterances are the typical way in which children in the material expressed emotions verbally. It might be possible to find similarities and differences among these utterances (e.g. whether they were assertives/directives/demands, etc.), but this was not pursued in the present study.
4.2.2 Emotive conventions
Conventional (albeit not descriptive) emotive expressions, i.e. utterances that are emotive through semantic conventions and thus less dependent on co-occurring nonverbal emotive expressions, appeared in the group with children from 3;2 years of age. One of the earliest – and most frequent – examples was dumma (‘stupid’ and its equivalents ‘silly’, ‘dummy’, etc. in English). Mostly the word occurred in constructions like dumma x (‘stupid x’), where x was a pronoun or a depreciatory noun. They are clearly emotive in nature, by definition rather than in actual use, and there are instances where the function appears to be evoking emotions rather than expressing them.
Two examples of early uses of conventional emotive expressions are illustrated below.
Klas (K, 3;2) and Richard (R, 2;1) at Klas’s house. The children have just entered the playing room after a short visit to the kitchen, where the mothers remain.
|1||[the boys are back at the railroad][Klas shoves Richard]|
|2||K:||dumma dej||you stupid||[Directed at Richard][Klas looks up at the researcher]|
|3||K:||får ja se, nu vill ja se||let me see, I wanna see now||[Said to researcher][Richard sits down and fiddles with two train carts]|
Klas seems to utter dumma dej (‘you stupid’) as a kind of explanation to the unprovoked pushing occurring seconds before. He then quickly glances up at the researcher, maybe to check whether the justification offered was accepted, and he changes the subject. This is a sequence where the prosodic traits are not easily detectable as emotive. Rather it is the “stupid x” in combination with the shoving that gives the emotive interpretation. The acts do not lead to any further aggressive interaction between the children, as Richard seems to disregard Klas’s behavior.
“Stupid poo” also stems from Klas in the same recording, but this time the prosody is clearly emotive and the target is his mother who is disciplining him. Klas is angry and upset since the mother has just given his toy car to Richard. Faced with a supreme power, he turns to language in order to retaliate.
Klas (K, 3;2) and his mother (C) in the kitchen at Klas’s home.
|1||C:||klas leker aldri me den dä de e bara om nån kommer näe klas vet du va du leker ju aldri med den om ja visste att de va din favoritbil då skulle du få ha den lyssna på mej nu=||Klas never plays with that, it’s just if someone comes and no Klas y’know what, you never play with it, if I knew it was your favorite car I would’ve let you have it, listen to me now=|
|2||K:||=DUMMA BAJSKORV||=STUPID POO|
Among the emotive conventional words children pick up early, we find “ouch”. It can be used with emotive prosody and while crying but it can also appear without these prosodic cues. In the example below, Klas (3;9) utters aj (‘ouch’) as a small plastic toy falls on his foot (out of view from the camera).
Klas (K, 3;9) in the play room together with his father (O) and the researcher (T).
|2||T:||oj fick du den på tån?||oh you got it on you toe?||[Referring to a small toy]|
|3||O:||kom ska jag blåsa.||come here I’ll blow on it.|
|4||K:||tuggummi kanske e bra?||maybe a chewing-gum is good?|
|5||O:||tuggummi? näe de tror ja inte //det brukar inte hjälpa//||chewing-gum? noo I don’t think so, //that usually doesn’t help//|
|6||K:||//kan vi inte prova//kan vi inte prova?||//can’t we try it// can’t we try it?|
|7||O:||nä. de hjälper inte.||noo. it doesn’t help.|
The way in which Klas utters “ouch” is clearly emotive, but as the interaction continues without any further prosodic cues of emotion, a possible interpretation is that the wording and prosody was used as a fixed multimodal expression (i.e. the prosodic contour of pain-calls used together with the conventional form of expressing “light” pain). The example illustrates that conventional emotive utterance can be useful in many ways, one being to aid children in handling emotionally complex situations; and another being to achieve other goals, like getting sympathy or, for that matter, to make a sibling receive a scolding.
Having used “convention” as a convenient label, a problematic twist now needs to be considered. “Conventional” is hard to exactly define (Planalp and Knie 2002). How frequently does an expression have to be used – and by how many individuals – in order to count as conventional? Are there other criteria than frequency and many individual users that should be considered? There is no obvious answer to these questions. Many words used by children are new, invented in the moment although they conform to the syntactic, morphological and phonotactic rules of the language. In this study these words have ended up among the conventional Accompanying utterances even though they might never have been uttered before. However, the semantic connotations of the parts used, together with the contexts of use, make them fit in the conventional emotive category. The example below is a typical illustration as it would also be an Accompanying utterance because of the simultaneous laughter.
The triplets are role-playing and using their Lego figures, letting them turn into various things. They laugh about both the utterances of the figures and the events the figures are encountering. The expression äckelströmme 1 ‘yuckifish’ appears together with laughter.
Disa (D, 5;5), Samuel (S, 5;5) and Molly (M, 5;5) in their baby sisters’ room.
|1||S:||de spelar ingen roll VA haru sönder mej? va haru sönder mej?||that doesn’t matter WHAT are you breaking me? what are you breaking me?||((Molly and Samuel start laughing))|
|3||S:||kan vi inte skiljas då?||can’t we get a divorce then?|
|4||M:||näe, ursäkta men ja e en äckelströmme||noo, excuse me but I’m a yuckifish||((laughing))|
Summing up, with regard to the verbal emotive expressions the children used where the semantics has emotive connotations, we have the following types of expression:
utterances carrying emotional semantic albeit not necessarily a conventional emotive-semantic form (“yuckifish”, “poohead”, etc.);
conventional words or utterances with emotive content (“stupid x”, “ouch”, etc.).
The conventional words related to emotions appear early, but after the prosodic one-word utterances from the first group of examples. Parents actively practice the use of some of these expressions, like the “ouch” followed by “I’ll blow on it” (i.e. make it heal/take away the pain). Oddly enough the “stupid x” seems to come from parents as well, as they use the expression when the children hurt themselves on objects, etc. (“what a stupid chair!”, “stupid stupid chair!”). That parents practice particular expressions with their children is known from earlier research as well (Clancy 1986). As a socialization force this is potentially very important, although not pursued further at this point. The earliest conventional words used in these data were “ouch” followed by “stupid” or “stupid x”. It appears that children start to use “stupid” concurrently with the two-word utterance “stupid x”, i.e. there was not a phase where they could only say “stupid” and then, as they grew a little more verbal, added “stupid x”. One hypothesis is that the scenes where these expressions appear are created by children close to or above 3 years of age. The cognitive understanding of conflicts, plots, and to maneuver with other people need to be present before the child understands the use of expressions like “stupid x”. A younger child – who in other situations can utter one or more words – will not use “stupid” or “stupid x” but rather cry and call for mommy when the going gets rough, regardless if s/he herself initiated a conflict scene or were the victim of one. They simply do not tune in and interact with the other children in the same manner and they do not use language as a tool in these interactions. This is important for the study of verbal emotive expressions because it implies that language development on its own will not account for the use of verbal expressions; one would also have to look at the incentive and social skills of the interacting children. Social interaction in a “typical” way takes time to learn and learning to display emotions in interaction takes even longer.
Among the Accompanying utterances there are one- and two-word expressions that constitute the first stumbling steps children take on the road to use language in situations where they previously only used vocal and nonverbal expressions. Potentially this is the group of utterances that later on disappears, but there is also the possibility that they continue to be used as verbal reactions connected to our physiological emotive reactions (akin to the quick and loud inhalation people do when startled). This is an empirical question and has to be addressed in the future. In this small-scale data corpus, the one-word prosodic utterances were present throughout, although not as frequent in the oldest age group (5;3–5;10), as the children were more likely to utter an “ouch” with emotive prosody/tears rather than a nonconventional one- or two-word exclamation.
As the children become more fluent they start to use longer phrases (counting three or more words and proper syntax) emotively. Furthermore, the children need to be emotively engaged and fluent at the same time. This group of expressions is likely to cover the entire lifespan. In the present data, children from the age of 2;1–2;11 started to use phrases in positive emotive situations, while the crying-and-speaking situations were not present until age group 3;2–4;0, indicating that the negative and positive emotive situations should be studied separately, as they might be connected to difficulties of different sorts for children (and potentially adults as well).
Descriptive emotive utterances have no necessary distinction between prosody and convention since they are all conventionally tied to emotions through semantics. Whether the descriptions reveal a real emotion (which might be discernible through prosodic traits, context, etc.) or describe someone else’s emotions, fictive emotions, potential emotions, and so on, is not important for the category. Descriptive utterances among adults might be more elaborated and in need of further analyses, but based on the present material of mainly child interaction this goal was not pursued. The children did not use Descriptive expressions of emotions on a daily basis. Nor did they use metaphors to this end. However, following them up through the ages we would probably see these forms start to appear. Frames like “I feel…” could be elicited on direct questions (e.g. “How did you feel…”) and potentially question–answer pairs like these constitute one of the ways in which children get accustomed to these verbal strategies to express and handle emotions. The example, with a father prompting and rehearsing “I love you” – “I love you too” with his 2;1-year-old daughter, indicates that another way to learn these expressions could be by practicing the contexts of use with more skilled language users (Gerholm 2007).
The fundamental difference between Descriptive and Accompanying utterances is that the descriptive utterances will be emotive whether the individual speaker means to or not, as the semantics works by itself. However, the person uttering the words need not be emotively engaged. The Accompanying utterances, on the other hand, appear emotive due to the prosodic and contextual traits, but the words could be used in a number of non-emotive fashions, too.
Illustrating the developmental path of emotive utterances that are captured in the descriptions above we have Figure 1, where examples from the text are inserted as illustrations.
A note of caution, however, is that it should not be read as if one form of utterance disappears and is replaced by another; rather, it is an additive development where a number of aspects (e.g. differences in personality, context and cultural belonging) will be decisive for how individuals use the different kinds of emotive language that are at their command.
The study set out to answer three questions. These are now returned to and discussed from a developmental perspective. The first question was: “What is the relationship between children’s nonverbal/vocal emotive expressions and language?”. Based on the material described and illustrated here, the suggested answer is that, as for the prosodic patterns, the characteristic sounds of vocal emotive expressions such as crying, laughter, etc., continue to be used when the vocalizations are discernible words and phrases, i.e. language. Many of the one-word utterances in the material, appearing in combination with emotive manifestation of a nonverbal and vocal kind, thus kept the characteristic sounds of, for example, shrieks or crying. We know from previous studies that emphasis, tone of voice, whispering, etc., all have an emotive impact on utterances (Weeks 1971; Papousek et al. 1990; Traunmüller and Eriksson 1993), but rather little about how these specific patterns develop or are chosen. Focusing on the emotive expressions only, we have one possible explanation since the vocal emotive expressions are inherent and present from birth (or, for laughter, developed within the first months of life, Wolff 1987; Masataka 2003). There is, potentially, a vast number of prosodic traits that mark an utterance as emotive and we would have to look further into the relation between somatic reactions and vocalizations in order to fill the descriptive void for the relationship between emotive vocalizations and language (cf. Turner 2000).
Turning to other nonverbal acts and their relation to emotive language, the results from the study indicate that young children expressing emotions in some nonverbal or vocal way also uttered one- or two-word utterances relating to the same emotive event. This has also been noted by Bloom (1993), who studied children in the ages 0;9 to 2;6. Bloom claimed that children in the beginning of language acquisition tend to verbalize after emotive expressions of a nonverbal kind, rather than simultaneously. Bloom further concluded that children continue to express their emotions nonverbally and that their verbalizations addressed what the emotions were about rather than the emotions themselves. This was so, said Bloom, since the mothers continued to interpret and respond to the children’s nonverbal displays in an accurate manner, i.e. the child had no incentive to leave the nonverbal mode and go verbal. What Bloom calls emotive verbalizations in the age group she studied would, in my terminology, be Accompanying emotive utterances. Children do not have to name the emotion, nor do they in a semantic-conventional way have to express emotions, but the utterance remains prosodically anchored to the child’s experience of emotions in the here and now. The early Accompanying utterances are in many ways similar to the nonverbal emotive expressions. Children begin to use verbalizations in situations where they previously have acted only nonverbally. Utterances containing onomatopoetic verbalizations, “no/not” and “mommy” appeared to be used in this way in the material. However, there are likely to be other words used in similar emotive ways, too, such as affectionate labels for significant others, exclamations, etc. As children concurrently are picking up the semantic conventions of how specific words are used, they will also start using these in circumstances that seem fit. This is the category of conventional accompanying utterances described in Section 4.2.2 where we find utterances like “silly”, “stupid poo”, etc.
The second question concerns: “What kind of verbal emotive expressions do children use as they interact in a home environment with parents and siblings?”. Continuing with the developmental path, the demands on children’s behavior increase as they enter the toddler period (Clancy 1986; Dunn and Munn 1986; Gerholm 2007). Thus, whereas children in the ages studied by Bloom might very well lack incentive to go verbal – if the purpose is to make oneself understood – older children and adults cannot rely on people interpreting them favorably as they throw objects, cry and scream. Instead, they are required to verbalize not only the causes, needs and desires they have but also the actual emotions. The Accompanying utterances were described as of two kinds, albeit sharing a fundamental structure of emotive prosody. The one kind of Accompanying utterance relies on the prosody and context to get an emotive interpretation, the specific wordings being less important (as when Samuel expresses sadness in ex. ). This description would also explain adult utterances that come through as emotive, although the words themselves state something else or even the opposite (“I’m not irritated, but…”). Another kind of Accompanying utterance is socialized ways of handling emotions. For the children many of these conventional situations are emotive in themselves but there are also examples where the conventional part is the heavier (as in the “ouch”-illustration, ex. ).
Turning to the Descriptive emotive utterances, they are candidates for being our cognitive path to emotional experiences in interaction. It is through labeling, descriptive and elaborated language, including emotion terms (like anger, sad, happy) and phrases (“I feel…”) that we manage to grasp, understand and handle our emotions in social interactions. Preverbal and young children cannot use language in this way, but the Accompanying conventional utterances might constitute the beginning of a development. Descriptive utterances in general would be a straightforward way of relating and dealing with emotions experienced since they actually bring them to the fore by labeling them. By elaborating the words, a process of distancing might further take place (Eisenberg and Mussen 1989). However, the Descriptive emotive utterances in the data corpus are few, implying either that the children have not yet started to use them regularly or that Swedish people in general do not topicalize (own) emotions and experiences by naming them. If the latter, children would not come across these utterances in the surrounding and would therefore not use them a lot themselves.
The third question concerns: “How do we go about studying verbal emotive expressions?”. It has been argued that this must be done in close proximity to the nonverbal and vocal expressions and the general context of the verbal acts. Interactional aspects need to be highlighted, as emotive expressions seldom take place in a social vacuum; on the contrary, they are responses and reactions to events in situ and develop in close relation to other people’s emotive expressions (Goodwin et al. 2012). Without contextual information an emotive utterance is not easily identified as emotive in a particular way. This means that the use of video recordings and phonetic measurements would be preferable. It has been argued that expressions leak through us in one way or the other (Lewis et al. 1993). If this is true we would expect individuals who do not express emotions verbally to be more prone to express them in other ways (increased heart rate, sweating, acting out physically, etc.). The findings stem from experiments on young children (Lewis et al. 1993) and it has, to my knowledge, not been investigated whether this holds outside of experimental settings or in the population as a whole. Given that the layers of emotive language that have been presented here are correct, or on the right track, we would have other, less simplistic, ways to describe and understand emotive expressions. People and cultural groups might differ in how up-front they are in revealing emotions verbally through descriptive utterances, but this would not imply that they would be more prone to use violence, etc. Instead, they might use a higher degree of Accompanying utterances. This is worth looking into further, as are the developmental aspects of emotive expressions. We would need to know whether the developmental path described based on 11 Swedish children is also valid for children in general. If there are differences, the ways the developmental path differs and in what speed the development takes place would give us important information on humans as a species, and the role of emotions and language of emotions in our common development and maturation. By using the categories described in this article, we have a means to divide utterances into developmental areas and compare these between cultures and cultural subgroups. We also have the possibility to describe changes appearing over time, in groups of people or in individuals, as social conventions, norms and common verbal expressions slowly change.
This study set out to describe the types of verbal emotive expressions that were used by children in their daily interactions with siblings, parents and peers. The examples were divided into two major categories depending on if the semantics used actually included emotion words (Descriptive utterances) or were emotive due to prosodic traits and context (Accompanying utterances). For the Accompanying utterances it was argued that the prosodic part kick-started a developmental process which later turned into the conventional utterances. This is an acquisition process where words and phrases are added piece-by-piece, leaning on cognitive growth, socio-emotional maturation and language acquisition in general. Descriptive utterances were rare and it is unclear if this depends on the input received or if Descriptive utterances of emotions are learned later, or perhaps belong to a different genre of speech than the multiparty family-life context provides.
Appendix. Key to transcription
underlining indicates emphasis
stretched sounds or syllables
marks nonverbal and contextual aspects; written in the fourth column
Anward, Jan. 1986. Emotive expressions. In Östen Dahl (ed.), Papers from the ninth Scandinavian conference of linguistics, Stockholm, Jan. 9–11, Stockholm: Dept. of Linguistics, University of Stockholm. Google Scholar
Bloom, Lois. 1993. The transition from infancy to language: Acquiring the power of expression. Cambridge: Cambridge University Press. Google Scholar
Bull, Peter. 2002. Communication under the microscope: The theory and practice of microanalysis. East Sussex & New York: Routledge. Google Scholar
Cauldwell, Richard. 2000. Where did the anger go? The role of context in interpreting emotions in speech. In Roddy Cowie, Ellen Douglas-Cowie & Marc Schröder (eds.), Proceedings of the ISCA workshop on speech and emotion (CD-ROM). Belfast, UK: International Speech Communication Association. Google Scholar
Cekaite, Asta. 2012. Affective stances in teacher-novice student interactions: Language, embodiment, and willingness to learn in a Swedish primary classroom. Language in Society 41. 641–670. CrossrefGoogle Scholar
Clancy, Patricia. 1986. The acquisition of communicative style in Japanese. In Bambi Schieffelin & Ellinor Ochs (eds.), Language socialization across cultures. Cambridge: Cambridge University Press. Google Scholar
Damasio, Antonio. 1999. The feeling of what happens: Body and emotion in the making of consciousness. New York: Harcourt. Google Scholar
Denham, Susanne. 1998. Emotional development in young children. New York: The Guilford Press. Google Scholar
Eisenberg, Nancy & Paul Henry Mussen. 1989. The roots of prosocial behavior in children. Cambridge: Cambridge University Press. Google Scholar
Ekman, Paul. 1989. The argument and evidence about universals in facial expressions of emotions. In Hugh Wagner & Antony Manstead (eds.), Handbook of social psychophysiology. New York: Wiley. Google Scholar
Ekman, Paul & Wallace V. Friesen. 1978. Facial action coding system. Palo Alto, CA: Consulting Psychologists Press. Google Scholar
Fridlund, Alan J. 1997. The new ethology of human facial expressions. In James Russell & José-Miguel Fernandez-Dols (eds.), The psychology of facial expressions. Paris: Cambridge University Press. Google Scholar
Gauvin, Mary & Susan M. Perez. 2006. The socialization of cognition. In Joan E. Grusec & Paul D. Hastings (eds.), Handbook of socialization, 588–613. New York: The Guilford Press. Google Scholar
Gerholm, Tove. 2007. Socialization of verbal and nonverbal emotive expression in young children. Dept. of Linguistics, Stockholm University Doctoral dissertation. Google Scholar
Gleason, Jean Berko & Sandra Weintraub. 1976. The acquisition of routines in child language. Papers and Reports on Child Language Development, 10. Stanford University. Google Scholar
Golub, Howard L. & Michael J. Corwin. 1985. A physioacoustic model of the infant cry. In Barry M. Lester & Zack Boukydis (eds.), Infant crying: Theoretical and research perspectives. New York: Plenum. Google Scholar
Goodwin, Majorie, Asta Cekaite & Charles Goodwin. 2012. Emotion as stance. In Marja-Leena Sorjonen & Anssi Perakyla (eds.), Emotion in Interaction, 16–41. Oxford: Oxford University Press. Google Scholar
Harrigan, Jinni A., Robert Rosenthal & Klaus Scherer (eds.). 2005. The new handbook of methods on Nonverbal Behavior Research. Oxford: Oxford University Press. Google Scholar
Howell, Signe. 1989. “To be angry is not to be human, but to be fearful if”: Chwong concepts of human nature. In Signe Howell & Roy Willis (eds.), Societies at peace: Anthropological perspectives. London & New York: Routledge. Google Scholar
Jacques, Sophie & Philip David Zelazo. 2005. On the possible roots of cognitive flexibility. In Bruce D. Homer & Catherine S. Tamis-LeMonda (eds.), The development of social cognition and communication, 53–81. New York: Psychology Press. Google Scholar
Juslin, Patrik N. & Petri Laukka. 2003. Communication of emotions in vocal expression and music performance: Different channels, same code?. Psychological Bulletin 129. 770–814. CrossrefGoogle Scholar
Kövecses, Zoltán. 2000. Metaphor and emotion: Language, culture, and body in human feeling. Cambridge & New York: Cambridge University Press. Google Scholar
Kuhn, Laura J, Michael T. Willoughby, Makeba Parramore Wilbourn, Lynne Verbob-Feagans & Clancy B. Blair. 2014. Early communicative gestures prospectively predict language development and executive function in early childhood. Child Development 85(5). 1898–1914. Google Scholar
Kurian, George. 1986. Parent-child interaction in transition. New York: Praeger/Greenwood Press. Google Scholar
Lewis, Michael. 2000. The emergence of human emotions. In Michael Lewis & Jeannette M. Haviland-Jones (eds.), The handbook of emotions, 2nd edn New York: The Guilford Press. Google Scholar
Lewis, Michael, Douglas S. Ramsay & Kiyobumi Kawakami. 1993. Affectivity and cortisol response differences between Japanese and American infants. Child Development 64. 1722–1731. CrossrefGoogle Scholar
Linell, Per. 2009. Rethinking language, mind and world dialogically. Charlotte, NC: Information Age Publishing, Inc. Google Scholar
Masataka, Nobuo. 2003. The onset of language. Cambridge & New York: Cambridge University Press. Google Scholar
Miller, Peggy & Linda L. Sperry. 1987. The socialization of anger and aggression. Merrill-Palmer Quarterly 33(1). 1–31. Google Scholar
Nelson, Judith K. 2005. Crying, caregiving, and connection: An attachment perspective. Florence, KY: Brunner-Routledge. Google Scholar
Ochs, Elinor. 1986. From feeling to grammar: A Samoan case study. In Bambi Schieffelin & Elinor Ochs (eds.), Language socialization across cultures, 251–272. Cambridge: Cambridge University Press. Google Scholar
Ochs, Elinor & Bambi Schieffelin. 1989. Language has a heart. Text 9(1). 7–25. Google Scholar
Oster, Harriet, Douglas Hegley & Linda Nagel. 1992. Adult judgments and fine-grained analysis of infant facial expressions: Testing the validity of a priori coding formulas. Developmental Psychology 28. 1115–1131. CrossrefGoogle Scholar
Panksepp, Jaak. 2000. Emotions as natural kinds within the mammalian brain. In Michael Lewis & Jeannette M. Haviland-Jones (eds.), Handbook of emotions, 2nd edn New York: The Guilford Press. Google Scholar
Papousek, Mechthold, Marc H. Bornstein, Chiara Nuzzo, Hanus Papousek & David Symmes. 1990. Infant responses to prototypical melodic contours in parental speech. Infant behavior and development 13. 539–545. CrossrefGoogle Scholar
Planalp, Sally. 1998. Communicating emotion in everyday life: Cues, channels, and processes. In Peter A. Andersen & Laura K. Guerrero (eds.), Handbook of communication and emotion, 29–48. New York: Academic Press. Google Scholar
Planalp, Sally & Karen Knie. 2002. Integrating verbal and nonverbal emotion(al) messages. In Susan R. Fussell (ed.), The verbal communication of emotions: Interdisciplinary perspectives. Mahwah, NJ & London: Lawrence Erlbaum Associates. Google Scholar
Psathas, George. 1995. Conversation analysis. The study of talk-in-interaction. London: Sage Publications. Google Scholar
Sandlund, Erica. 2004. Feeling by doing: The social organization of everyday emotions in academic talk-in-interaction. Doctoral dissertation. Karlstad: Karlstad University Studies. Google Scholar
Schieffelin, Bambi. 1979. Getting it together: An ethnographic approach to the study of the development of communicative competence. In Elinor Ochs & Bambi Schieffelin (eds.), Developmental pragmatics. New York: Academic Press. Google Scholar
Schieffelin, Bambi. 1986. Teasing and shaming in Kaluli children’s interactions. In Bambi Schieffelin & Elinor Ochs (eds.), Language socialization across cultures. New York: Cambridge University Press. Google Scholar
Shaw, George B. 2004. A treatise on parents and children, 1st edn. Fairfield, IA: World library – Literary Society. Google Scholar
Shields, Maureen. 1979. The child as psychologist: Constructing the social world. In Andrew Locke (ed.), Action, gesture and symbol: The emergence of language. London: Academic Press. Google Scholar
Thelen, Esther & Linda B. Smith. 1994. A dynamic system approach to the development of cognition and action. London: MIT Press. Google Scholar
Thompson, William F. & Laura-Lee Balkwill. 2006. Decoding speech prosody in five languages. Semiotica 158(1/4). 407–424. Google Scholar
Traunmüller, Hartmut & Anders Eriksson. 1993. F0-excursions in speech and their perceptual evaluation as evidenced in liveliness estimations. In Experiments in speech processes, PERILUS No. XVII, 1–34. Department of Linguistics. Stockholm: Stockholm University. Google Scholar
Turner, Jonathan H. 2000. On the origins of human emotions: A sociological inquiry into the evolution of human affect. Stanford, CA: Stanford University Press. Google Scholar
Vygotsky, Lev S. 1978. Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press. Google Scholar
Wittgenstein, Ludwig. 1953. Philosophical investigations. Oxford: Blackwell. Google Scholar
Wolff, Peter H. 1987. The development of behavioral states and the expression of emotions in early infancy. Chicago: The University of Chicago Press. Google Scholar
About the article
Tove Gerholm is Senior Lecturer at the Department of Linguistics, Stockholm University. Her main research interests are multimodal communication and the development of language and communication in children. At present she is investigating these aspects in two projects: (i) the MINT project (http://su.avedas.com/converis/project/218) and (ii) enhancing preschool children’s language skills (http://www.buv.su.se/english/research/research-projects/early-childhood-education/enhancing-preschool-children-s-attention-language-and-communication-skills-1.209094?cache=title%3Duse).
Published Online: 2018-02-27
Published in Print: 2018-02-23
Citation Information: Text & Talk, Volume 38, Issue 2, Pages 137–165, ISSN (Online) 1860-7349, ISSN (Print) 1860-7330, DOI: https://doi.org/10.1515/text-2017-0036.
© 2018 Gerholm, published by De Gruyter. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. BY-NC-ND 4.0