A constituição prosódica da enunciação na relação mãe-bebê

RESUMO No presente estudo de casos, discutimos a produção de sentidos carregados com elementos prosódicos no começo da vida. Alinhamos essa discussão com observações de Bakhtin acerca de características da enunciação. Com esse alinhamento investimos em explicações sobre o status linguístico da prosódia nos diálogos entre adulto e bebês em situação de aquisição de linguagem. Os dados foram registros videográficos da interação de duas díades adulto-criança, uma francesa e uma brasileira, durante atividades cotidianas1. Nas análises, comparamos variações da curva entonacional dos enunciados dos adultos com vocalizações dos bebês. Nos resultados discutimos como essas variações refletiram exercícios de posições axiológicas, discutidas no âmbito da apropriação enunciativa. Concluímos que os aspectos prosódicos são recursos principais para enunciação no começo da vida e vinculam as práticas com a linguagem ao desenvolvimento humano.


Introduction
In this study, we defend the idea that the prosodic aspects in speech at the beginning of life have a linguistic function. We anchor our argument in the discussions proposed by Bakhtin (1986) 2 about enunciative appropriation and the broad dialogical relationships that link language practices to human development.
The discussions about enunciation were the main milestone in Bakhtin's work as he contests traditional procedures practiced by Linguistics and Cognitive Psychology, since enunciation ascendsfor Bakhtinto the role of voice, which is another innovation in the context of human language proposed by the author. We consider the advances in the way the uses of language are conceivedand, above all, the depth of the author's explanations about the interrelationship between enunciation and expressive intonationto discuss suprasegmental elements of speech at the beginning of life. In our discussion, we review Bakhtin's and Vološinov's (VOLOŠINOV, 1973)  According to this article's approach, the baby is recognized as a conversational partner by an adult who addresses him or her using language with specific characteristics. This language is designated by various terms, such as inputa term used by different authors and theoretical perspectives to refer to linguistic and non-linguistic stimuli received by the baby -, as well as "Infant Directed Speech (IDS)" (FERNALD, 1994) and Child Directed Speech (CDS) (SNOW, 1979), and also by the term "motherese (PAPOUSEK; PAPOUSEK; SYMMES, 1991). The linguistic characteristics that adults use when talking to babies show lexical, grammatical and prosodic specificities, differentiating this language from the one that circulates among adults. Fernald (1994) pointed out that when adults communicate with babies, there is a predominance of similar linguistic characteristics that can be found across cultures (although this language may also be permeated by cultural variability). However, this position is counter-argued by 2  researchers who base their discussions on interactionist perspectives, which emphasize the predominance of variability over similarity as a resource for the socialization of cultural characteristics (SNOW, 1997).
Although Bakhtin (1986) 4 does not directly refer to the moment of language acquisition, the author states that words are not acquired from dictionaries, with complete and closed meanings. For Bakhtin, words acquire meaning during communication, in the unique enunciative contexts in which they are used (FARACO, 2003). Therefore, the acquisition process results from real conversational exchanges with utterances that perform different voices/positions in the interaction. In the utterances, the words of the other, directed at the child, are gradually assimilated, becoming the child's own words (VOLOŠINOV, 1973;5 VOLOŠINOV, 1983). 6 In this dynamic, it is the intonation that marks the function of otherness, which emerges with the other's word, according to Bakhtin. Child development is replete with the assimilation of the words of others (with otherness). Throughout the process of assimilation, those words are re-emphasized, imprinting the axiological position (attribution of value) of the speaker, who thus transforms them into words of their own. Bakhtin points out that: This is why the unique speech experience of each individual is shaped and developed in continuous and constant interaction with other's individual utterances. This experience can be characterized to some degree as the process of assimilationmore or less creative -of other's words (and not the words of a language). Our speech, that is, all our utterances (including creative works), is filled with other's words, varying degrees of otherness or varying degrees of "our-own-ness," varying degrees of awareness and detachment. These words of others carry with them their own expression, their own evaluative tone, which we assimilate, rework, and re-accentuate (1986, p.89). 7 According to this interpretation, in the initial moments of child development, the concrete statements, and specially the voice of the children's interlocutors, are the marks of the speaker's subjectivity oriented towards the otherthe child (DAHLET, 2005;4 For reference, see footnote 2. 5 For reference, see footnote 3. 6 VOLOŠINOV, V. N. The Construction of the Utterance. In: SHUKMAN, Ann (ed.). Bakhtin School Papers. Russian Poetics Translation, Vol. 10. Trad. Noel Owen. Somerton: Old School House, 1983. pp.114-138. 7 For reference, see footnote 2. BERTAU; GONÇALVES; RAGGATT, 2013;VOLOŠINOV, 1983). 8 Thus, as the child faces these words of others, they also face the otherness implied in the particular position of the one who speaks.
We observe that, in order to become one's own words, the words of others need to be materialized in the linguistic sign. We then argue that prosody is the main resource that makes this materialization possible, given its prevalence in communication at the beginning of life. In other words, in the early stages of child development, prosodic resources enable children to assimilate the voices/positions of their interlocutors who, through dialectical relationships, become part of their own unique and particular voice/position. According to Hermans (1996;2001) and Silva and Vasconcelos (2013), this multiplicity of voices/(ideological) positions in permanent dialogue characterizes the dialogical self. In this approach, a person is in constant change, since the voices (positions) that constitute them reflect conflicts, agreements and disagreements We evaluate this broad discussion on the constitutive aspects of the language acquisition process and its impact on child development in order to present and explore information about a research work that aimed to analyze the organization of prosody in adult-baby interaction, intending to situate this organization in the discussions about the enunciative appropriation (BAKHTIN, 1984). 9 Our expectation was to gather clarifications on the linguistic dimension underlying the way prosody works in early life.

Prosody: Concept and Operation
In the work of the Bakhtin Circle, it is possible to find references to the term intonation, and more specifically, expressive intonation, as to characterize the evaluative/appreciative orientation (ideological orientation) in the enunciation as one of the aspects that highlights the dialogical dimension that sustains the relationship between language and human development. However, we must warn that the term intonation is used, from a linguistic point of view, to comprise a variety of elements, currently defined as prosodic, paralinguistic and extralinguistic. Nowadays, the term prosody refers to all All content of Bakhtiniana. Revista de Estudos do Discurso is licensed under a Creative Commons attribution-type CC-BY 4.0 musical aspects of language: a field dedicated to the study of the constituents of oral expression (such as stress and tones) manifested through variations in fundamental frequency (F0), duration and intensity, perceived as changes in pitch and duration, for example.
In the literature, authors like Crystal (1976) contrast prosodic and paralinguistic traits. According to this point of view, prosodic elements are considered to be the most conventional and systematized elements of language and, therefore, "more linguistic" than those considered paralinguistic. The paralinguistic elements, on the other hand, include voice quantifiers (whispered voice, resonant, creaking, falsetto) and voice qualifications (laughter, tremor, crying), which would not have the same degree of systematization in the language as the prosodic ones, being therefore considered "less linguistic." Following this theoretical proposal, Vasconcelos and Leitão (2016) observed that although the children's cry may be categorized as extralinguistic, mothers interpret it as completely meaningful since the beginning of the child's development. Mothers attribute meaning to their children's crying based on linguistic elements, differentiating, for example, 'crying' from 'weeping', based on differences in pitch and duration of these two actions (crying is normally produced with higher pitch and longer duration). It is relevant to point out that this distinction is based on prosodic criteria (especiallybut not exclusivelyin duration and intensity). That leads to the understanding of prosody as a basis for systematizing linguistic criteria used to differentiate pitch and duration in children's crying. We then reinforce questionings about the validity of classifying and distinguishing elements as more or less linguistic. We point out to the various sound continuum modulations (in pitch, duration and intensity), which are polissemic processes and also convey linguistic information that is relevant to the comprehension and interpretation of utterances. The utterances also carry intersubjectivity, pragmatics, etc.
The distinct voice qualities in the speech addressed to the child (nasal, whispered, tearful, pleading, murmured, etc.) produce distinct effects on the meanings that are being conveyed. The use of falsetto (which implies the choice of a high pitch) by the adult, for example, often has the purpose of attracting and keeping the child's attention. The discursive function falsetto assumes in interaction is revealed by the frequent use of this voice quality to differentiate moments in which the adult speaks for the child and occupies the child's discursive position in the dialogue (position marked by the use of falsetto) from the moments when the adult speaks for themselves, addressing the child as an adult (and therefore not using falsetto). The prosodic characteristics underlying this voice quality (falsetto) directed to the child include the use of broad intonation contours, slow speech with repetitions, high fundamental frequency, greater number of secondary stress, among others. This prosodic variation imprints an analytical social function to the speech directed to the child and delimits its linguistic units (SNOW, 1997;GARNICA, 1979).
According to Cavalcante (1999), the melodic distinctions of maternal productions insert the child into the language, especially during their first year of life, in which the prosodic elements of maternal speech play a discursive role. Thus, according to the author, the mother uses the changes in her voice to create discursive spaces that insert the baby into the interaction, and she gradually expands these spaces, supporting the development of the baby's unique position in discourse. With regard to the language acquisition process, it is possible to state that the adults assign a discursive place to the child through their own voices. Therefore, adultsmore often the mother -"give" their voices to the child, delimitating the moments when they speak for the child by the use of suprasegmental elements (such as the falsetto voice). The child then appropriates the words of others and makes use of them. This use, however, does not invalidate the marks of otherness (the quality of voice and the dialogue with adults). We assume that this dynamics of appropriation and use of other people's words with the practice of otherness at the beginning of life refers to the appropriation of the enunciative positions discussed by Bakhtin (1986). 10 Thus, in this paper and as pointed out before, we aim to analyze the organization of prosodic aspects in interactions of two adult-child dyads, in order to clarify in which extent this organization results from the definition of the children's enunciative positions created by adults. With this focus in mind, we make use of information that leads us to the understanding of the linguistic function of prosody.

Enunciation and Utterance
According to the Bakhtin Circle, the enunciation/the utterance's organization assumes a prominent role when looking for the fundamentals and explanations about the dialogical functioning of linguistic/discursive operations. The Bakhtinian standpoint considers enunciation as the living object of communication. Thus, it criticizes the normative description of language, which separates it from its concrete production contexts. In Bakhtin's linguistics (1986), 11 the utterance is the "real unit of speech communication" (p.67). According to this perspective, the chosen object of study in linguistics is enunciation (the utterance), understood as the speaker's discursive performance.
One of the first characteristics of enunciation that can be explored is the place/time of authorship (as opposed to the study of abstract language units that do not belong to anyone). In the Bakhtinian conception of enunciation, the authorship position does not relate to subjectivist or individualistic points of view, nor does it refer to the experience of the physical author (the author-person) who speaks. In this approach, authorship refers rather to the discursive position (of the author-creator) that presupposes a historically and socially situated experience of those who assume it. To reinforce his explanations, Bakhtin warns that dialogic relationships are reducible neither to logical relationships nor to relationships oriented semantically toward their referential object, relationships in and of themselves devoid of any dialogic element. They must clothe themselves in discourse, become utterances, become the positions of various subjects expressed in discourse, in order that dialogic relationships might arise among them (1984, p.183). 12 It is then observed that in each enunciative situation, even if it is composed of only one word, a specific completion takes place, and it includes the alignment of authorial positions and enables the emergence of responses (responsiveness). According to the author, this kind of relationship is impossible to be found between the units of the language alone; it is rather a property of enunciation, as enunciation promotes the negotiation of discursive positions. In this dynamic, the possibility to respond (to have a responsive attitude) is the most important condition for the completion to happen. Therefore, it is not enough for the statement to be understandable at the level of the sentence, or to retain its linguistic clarity; it is only through the completion aspect that the reaction, a characteristic of the enunciation itself, is revealed. In other words, the responsiveness underlying enunciation ensures the dialectical purpose that an authorship position holds (BAKHTIN, 1986). 13 To summarize, in the present research we approach the concepts of enunciation and responsiveness presented by Bakhtin in order to convey the relationship between those concepts and some current interpretations of prosody (variations in fundamental frequency, duration and intensity) in the adult-baby interaction. We focus on the organization of prosodic aspects in adult speech related to the baby's productions, so that it is possible to clarify that this relationship goes back to the enunciative appropriation processes discussed by Bakhtin (1986), 14 in which the link between the uses of language and human development can be detected.

Methodology
This research aimed to analyze the organization of prosody in adult-baby interaction, intending to situate it in the discussions about enunciative appropriation (BAKHTIN, 1986). 15 Our expectation is to recognize that prosody plays a linguisticdiscursive role since the baby's first months of development (CAVALCANTE, 1999;VASCONCELOS, 2017). This article is defined as a case study, based on an idiographic approach, characterized by the systemic analysis of phenomena in their variability and context (SATO et al., 2007). It is assumed here that knowledge is a constructive-interpretative production that results from a cyclical process, through which the researcher's world concepts and experiences, the phenomenon itself, the method and the theory are integrated (BRANCO; VALSINER, 1997).
In order to discuss our premises, we longitudinally observed two children (one male and one female). The male child is a monolingual Brazilian child. His data were All content of Bakhtiniana. Revista de Estudos do Discurso is licensed under a Creative Commons attribution-type CC-BY 4.0 recorded in monthly videography sessions (with recordings varying from 30 minutes up to 1 hour) of daily interaction with his family members (during meals, bathing and playing), throughout his first thirty-two months of life (from the age of four weeks to the age of two years and eight months). For the analyses, we chose to delimit the period between his sixth and thirty-second months of age, between the years of 2012 and 2014.
As an effort to protect the ethical procedures for research with human beings, the observed child was named V., the only son of a middle class family in northeast Brazil.
In addition to V., the other participant of this study is his mother, who interacted with him during the sessions.
The female child (named M. for the purposes of this study) was a French monolingual, member of a middle class family from Paris, France. M.'s recordings were also made on a monthly basis (in this case, from the age of eleven months, from 2006 to 2008). In the analyses, we delimited the discussion data up to her thirty-second month of age. In M.'s case, each recording has the duration of an hour and also takes place in her daily life situations (these data belong to the Colaje group and are available on the CHILDES platform). 16

Procedures for Data Organization and Analysis
The data transcriptions were performed with the CLAN-program (data from M. can be accessed on the CHILDES platform in French). The recordings of the Brazilian child were transcribed using those same parameters. Each transcript contains different transcription lines: the ones indicated by initials in English correspond to the orthographic transcription of speech turns (CHI, child; MOT, mother). When there was no speech in the child's turn (characterized as a non-verbal turn), it was represented by 0 (zero), and the transcript then only presents the description of the action performed by the child, the letter 'y' in repetitionyyy represents the babbling of the baby. The other transcription lines refer to the description added by the transcriber, according to the scene (% act, which describes the actions performed by the adult or the child; % sit, which describes the interactive situation that takes place in the scene; % com, which adds the researcher's comments). The acoustic analyses were conducted with the software PRAAT. Each 16 http://childes.psy.cmu.edu/data/Romance/French/ production was syllabically segmented. The statements' intensity peak and the pitch directions in the fundamental frequency curve (F0) were also registered.
In the analyses of the vocal productions of the observed children we chose to describe the direction of the pitch curve, that is, how the children developed upward or downward pitch movements in their speech turns. Through the variation of these movements, basic intonational contrasts (ascending, descending, ascending-descending, descending-ascending, leveled) were configured, as well as intensity contrasts with higher or lower amplitude, pitch, voice quality and duration. Investigating these contrasts helped us analyze the meanings produced in the interactions we studied.
In regard to the transcription of the prosodic characteristics found in the analyzed statements, we opted for a simple system, which consisted of noting the initial, maximum, minimum and final points of the F0 curve. In the notes, we classify these points as: a) H (highest point or points on the pitch curve); b) L (point or points at the lower limit of the F0 boundary). We emphasize that this transcription system was used as a registration means and not as a phonological-prosodic tone classification. In an attempt to characterize the movements of melodic ascension and decline, the values of Inter F0 in semitones were added to the transcripts, such as HL (1.6), where the value 1.6 represents the difference of 1.6 semitones between the highest and lowest points in the pitch curve.
We also noted the duration of vocal production in milliseconds.
In summary, we focused our analysis on the variations of suprasegmental aspects such as pitch, intonation, voice quality and intensity (all prosodic features) to support -Bakhtin's discussions (1986) 17 about enunciative appropriation, which states the main argument about the relationship between the uses of language and human development.
For this purpose, we use empirical data from early communication.
The two children observed in this study had no linguistic-cognitive impairment.
It is important to emphasize that our analyses did not aim to compare the two children's developmental characteristics or rhythms. Thus, each child was analyzed individually, preserving the specificities of the two languages being acquired by each of them: Portuguese and French. Comparative analyses were only carried out considering different developmental stages of the same child; that is, comparing excerpts from videos collected in the initial phase of the selected data with excerpts from videos collected in a later phase of the same child. Analyzing two cases was a research strategy to favor the enrichment of the corpus and to provide the study with a greater contingent of data variability.
The analysis of two children who go through the acquisition process of two different languages (Brazilian Portuguese and French) was also a strategy to increase the variability of the study data. French and Portuguese integrate the set of Indo-European languages, which share some similarities and, at the same time, melodic particularities, such as, for example, distinct rhythmic and stress characteristics. The French child's data were selected given the procedures used for their collection and registration, which were identical to those used for the Brazilian child's data.
We also inform that other approaches in the analysis of these cases have already been the subject of discussion and previous publications with different focuses (VASCONCELOS; SCARPA; DODANE, 2018;2019). In this article, we explored situations that caught our attention due to their meaning production scenarios, which are particularly configured based on prosodic traits. More specifically, we explored four situations in which prosodic features proved to be central to meaning production in the interactions, at different children developmental stages. The data discussion was linked to the researchers' narrative analysis, in line with Bruner's (1991) and Flick's (2007) ideas about the narrative construction of reality. The attention drawn to the narrative in this investigation gives this study a qualitative perspective in regards to the data analysis.

Results and Discussions
All recorded episodes were transcribed in full. However, for the purposes of the present article, we selected four episodes (two involving the Brazilian child and two involving the French child) considering their concentration of situations marked by exchanges based on prosodic traits. Children's productions found in those excerpts were analyzed with the software PRAAT.
According to the objective of this research, we analyzed the organization of prosodic traits in the interaction of two adult-baby dyads in order to support Bakhtin's discussions about enunciative appropriation, in this case, with empirical communication data from the beginning of life. In other words, we approach the relationship between prosody and meaning negotiation in the interaction, which are related to the processes of All content of Bakhtiniana. Revista de Estudos do Discurso is licensed under a Creative Commons attribution-type CC-BY 4.0 language acquisition. Placing the locus of this research in the period of early language acquisitionnormally characterized as pre-linguistic period (due to the absence of words in children's speech)favored the attention to the variations in voice quality (pitch, duration and intensity), in which possibilities of meaning construction in interaction were imprinted, as we could observe in V. and M. cases.
In Episode 1 (V.'s case), V.'s mother pushed an object away from the child, with the objective of making the boy crawl until he could reach it. As V. was unsuccessful in his attempts to pick up the toy, he stretched his arms, raised his body and neck, looked at the toy and vocalized (Fig. 1). These actions were interpreted by his mother as a protest (see transcription below): Episode 01 -Look who is talking and complaining! 6 months 1.
%act: CHI looks at the toy and stretches the arm while swinging the body 3.
towards it.
CHI: 0. 6. %act: CHI looks at the toy while stretching the arm e hitting the ground 7.
%act: CHI looks to the side to another toy, moves the body towards 10.
the toy and vocalizes 11.
%act: CHI stretches the body and touches a toy, pushing it further away from him 14.
MOT  20 We note that each production has specific pitch, duration, intensity, rhythm and direction characteristics. However, for the purposes of our analysis, each transcript describes only the prosodic elements that seemed to be at stake in meaning production according to each context.
%act: CHI stops vocalizing and looks at the mother and lies on the ground Image 1. V. tries to reach the object, elevates the body, looks at the toy and vocalizes V.'s mother interpreted that the child was unhappy with the situation (say I'm not enjoying this no -line 14). We observe how the adult, when dealing with the child's actions and babbling, engages in attempts to interpret these signs, giving meaning to children's productions. Considering this, we also note that the child's vocalization, despite displaying no words from the adult language, revealed high pitch, intensity and duration characteristics, which enabled the mother's interpretation for his babbling as an expression of protest. In this case, the child's actions (trying to reach a toy, staring at it with a certain body tension) combined with a vocalization with specific prosodic traits enabled the mother to assign the meaning of protest or complaint to the child.
Regarding the prosodic characteristics of the child's productions in this episode, the average maximum F0 was 472 Hz, and the average initial F0 was 407 Hz, which can be considered generally high values, according to what is expected in children's productions. As a frame of comparison, we point out that the average child F0 found in a study with 182 Brazilian children in São Paulo state was 237.15 Hz (VANZELLA, 2006).
As the average melodic rise was between 2 to 4 semitones, the child's productions in the analyzed episode can therefore be characterized as vocalizations with high pitch and duration.
We also highlight the way in which the mother 'organized', in one sentence, her interpretation of the child's expression. Based on the child's actions and vocalization, the mother gave the child's production a finished aspect as she attributed a meaning (of protest) to it. Our attention to the child's active role in the interaction is relevant, since it is according to their productions that the adult builds an interpretation and communicates with the child. The child's action is necessary so that the adult can attribute a meaning to it and the dyad can therefore get involved in a communicative situation.
In this analysis, the mother's sentence say I'm not enjoying this no illustrates how the enunciation was marked by prosodic traits. Although it is a sentence with only one grammatical subject, it enabled two enunciative positions. That is, the verb say refers to the mother as a speaker and, at the same time, I'm not enjoying it, it refers to a possible speech coming from the child. We claim, then, that the two enunciative positions were sustained in this episode when the negotiation of two distinct axiological positionsthe one of the mother and the one of the childtook place.
In Episode 2 (M. case), we explore another example of the organization of aspects of children's utterance (bodily actions, gaze direction and, especially, babbling) activated in the mothers' utterances. In this episode, M. also does not yet produce words from her language; despite that, the specific melodic contour of her babbling was interpreted by her mother as a call, when she moved away and heard the child babbling at the stair foot (Image 2; see transcript below).
Episode 02 -Look who is talking and calling me! 10 months 1.

4.
%act: leaning on the stairs, looks up, where her mother is. 5.
CHI: vocalizes with descending intonation Image 2. M. follows her mother, who went upstairs, and vocalizes, a production which was interpreted as a call; on the right, the spectrogram that shows the upward contour of M.'s vocalization. The main purpose of this research was to clarify the role of prosodic aspects in the enunciative appropriation, considering communicative scenarios for language acquisition. With the example discussed in episode 3, we highlight the similarity between the prosody curves produced by the child (through babbling) and by the mother (through speech). M.'s mother reproduced the same prosody curve the child did (Fig. 4), 'filling it in' with words. In the present analysis, this similarity illustrated a scenario of enunciative appropriation, supported mainly by the organization of prosodic traits: the mother's interpretation of the child's babbling as a question on the inadequacy of the object's (cloth) place was based on the final ascending tone used by the child, a typical prosody of interrogative utterances in the French language. Thus, the prosodic aspect was fundamental in this scenario, as the mother reproduced the same melodic parameter, indicating responsiveness and, at the same time, expanding the enunciative situation by filling in the words missing in the child's turn. By expanding it, the mother added her values (with the language, for example), also illustrating the practice of otherness in the meaning. Through this meaning emergence, mother and child shared the transformation that took place in the scene, changing from a sense of 'warning' to a sense of humor, which could be confirmed by the laughter that invades the end of the dialogue.
In Bakhtin's words, "in all its various routes toward the object, in all its directions, the word encounters an alien word and cannot help encountering it in a living, tensionfilled interaction" (BAKHTIN, 1981, p.279). 27 Corroborating these words, we interpret that, in episode 4, the integration movement was configured from the responsiveness that characterizes the dialogue situation. In responsiveness, the child used the same tools found in his mother's speech (the speech of other). The laughter then resulted from the axiological position taken by the child, as his scarce ability with the correct expression of words led him to the production of a non-existing word: "castau." In turn, the laughter also affected the mother's axiological position, as she recognizes her child's scarce ability with words. The mark of the speech of the other is confirmed in the preservation of the similarity between the speech of the mother and that of the child, as the expected word ("castigo"/to be grounded) appears, in addition to the descending intonation, which indicates a sense of complementation to the unfinished utterance.
This scenario had a humorous effect for the mother. The humorous effect of this enunciative appropriation can be analyzed based on Bakhtin's (1981) 28 considerations about intertext and parody. According to the author, prosody is carnivalesque, ambivalent, bivocal and dialogical. In prosody, the serious voice is mocked, the discourse of authority is denied and the relativity of things is affirmed. This dichotomy is present in the analyzed episodethe naivety of the child's speech versus the seriousness of the mother's warningand it generates the parody effect and, therefore, the laughter. It was the linguistic marks of this production, both syllabic and intonation, that enabled the interpretation constructed here, according to which the child's statement presents bivocality. Thus, the prosodic characteristics of children's babbling were, once again, fundamental to the production of meaning. Therefore, we reinforce our argument about the definitely linguistic status of prosody, as it carries a symbolic function, which is constitutive of enunciative appropriation in the beginning of life. 27

Final Considerations
In this study, we based our considerations on the Baktinian concept of enunciation in order to defend the linguistic status of prosody in early life. We discussed the organization of prosody in dialogues in two adult-baby dyads. We argue that prosody is fundamental to the production of emerging meanings in communicative situations without the predominance of words. In the episodes analyzed in this research, we observe how the adult deals with the child's actions and babblings and engages in attempts to transform this signal into a sign, using prosodic elements to give meaning to children's productions. Our analyses confirmed that prosody is one of the first linguistic phenomena that can be observed in children's productions and it proved to be one of the first linguistic aspects for correspondence between partners in interaction.
Bakhtin's observations on enunciative appropriation enabled us to recognize prosodic aspects as very important in the period of language acquisition, as the babies' vocalizations, even when still without words, fostered meanings in the interaction with the adult. That happens because the enunciation's constitutive dialogical relations bound the exercise of axiological positions to the communicative process, visualized in our data in the similarity between intonational curves of the adult speech with those of the baby's babbling during dialogue.
In our analysis, we also point out to the way that the child in the process of language acquisition exercises its axiological position, guided by the words of others (by the discourse of others). As for the adults, we emphasize that this exercise took place when they filled the prosodic curve of babies' vocalizations with words. In this operation, prosody enables the existence of a sense of materialization in the dialogical constitution of language. The adult interpretation, based on prosodic elements, offers a sense of completion to the children's utterances.
We conclude, then, that the prosodic aspects are fundamental resources for the enunciative appropriation in dialogue with no predominance of words, which are typical in the beginning of life. Furthermore, our focus on the conception of enunciative appropriation helped to clarify that the linguistic status of prosody at the beginning of life resides on the fact that prosody is a fundamental link binding linguistic processes to the continuum of human development. In conclusion, we defend that the relationship between