1. Introduction

1Listing practice is an activity intended to elaborate, explain, illustrate or evaluate a sequence of speech (Selting, 2007). It requires a multi-unit turn produced by a single speaker but a few studies within the conversation analysis framework have described lists as a collaborative achievement by all the participants (Jefferson, 1990; Lerner, 1994). The current paper aims to investigate such collaborative work in listing practice. Using the notions of alignment and affiliation (Stivers, 2008), respectively defined as an adaptation to the activity in progress and an endorsement of the speaker’s stance, we then attempt to demonstrate that collaborative work by the recipient is not sufficient to make what we have called a “convergent sequence” (Guardiola & Bertrand, 2013). In listing practice, the recipient can add a potential list item that we have characterized as a specific feedback response – henceforth SFR – (Bavelas et al., 2000) in the current work. Convergent sequences require that the SFR be not only accepted but also ratified by the prior speaker. Based on a corpus of French dyadic conversations and drawing on various approaches, both in linguistics and psychology, focusing on conversations, we investigated a collection of lists including SFRs and the different interactional trajectories occurring after them. By focusing on the next development in the speaker’s speech, we try to fill the existing gap in studies on the ratification of feedback utterances. To date, indeed, feedback responses have been mainly investigated as reactive tokens. Following Tolins and Fox Tree (2014), we confirm the proactive nature of feedback. In this paper, we firstly focus on relevant features which make list construction a good candidate for illustrating a collaborative sequence. Secondly, we analyze the entire structure of lists (including ratification) related not only to the larger activity (i.e. storytelling here) in which they are embedded (Selting, 2007) but also to the sequences that lists contribute to constructing (word search, humor). We then show that a collaborative achievement does not necessarily achieve a convergent sequence as defined in (Guardiola & Bertrand, 2013).

2. Overview

2A large body of work on interactional achievement (Schegloff, 1982), elsewhere joint activity (Clark, 1996), in talk-in-interaction has focused on the collaborative interactional nature of conversations.

3Some studies in the conversation analysis framework have investigated collaborative work by examining collaborative turn sequences. Following Lerner (2004: 229): “A collaborative turn sequence is a collaboration of two speakers producing a single syntactic unit, not only in that a next speaker produces the completion to TCU [turn constructional unit] begun by a prior speaker, and that prior speaker does not continue once the pre-emptive completion begins, but also in that the first speaker ratifies the completion after its occurrence as an adequate rendition of the completion of the TCU they were about to voice”. Lerner thus considers that a collaborative turn sequence is composed of 3 components: a preliminary component (completion source or unfinished first turn), a pre-emptive completion and a receipt slot “in which the original speaker ordinarily reasserts authority over the turn’s talk by responding to the proffered completion (or by producing an alternative to it)” (Lerner, 2004: 225). Although this three-component sequence has been highlighted, very little work has been done on this last component (see however Mondada, 1999, and Oloff, 2014, for French). Only a few authors who have examined lists as a completion by the recipient, Jefferson (1990) under the terms of additive assimilation or Lerner (1994) under those of anticipatory completion, have considered this ratification by the prior speaker of the recipient’s completion. By paying attention to the interactional trajectories deployed through ratification by the prior speaker in listing practice, the present paper is an attempt to fill the gap of studies in the entire sequential organization of listing practice.

4Beyond the sequence level, listing practice can also be investigated with regard to activity. In this respect, one of the most widely studied activities, in various fields, is storytelling (see Schegloff, 1997; Norrick, 2000; Bavelas et al., 2000; Stivers, 2008; Guardiola & Bertrand, 2013). Storytelling is seen as an asymmetrical activity involving different discursive roles by the storyteller/narrator and the listener. The main results from different studies show an active collaboration by all the participants because the role of the listener is as important as that of the narrator for the successful achievement of storytelling. Specifically, Bavelas et al. (2000) showed that the listener’s role is to provide appropriate feedback responses. Appropriate responses, including verbal items such as “mm”, “yeah”, or gestural ones, were classified as generic when they display an understanding function while feedback displaying a more evaluative function was classified as specific (referred to as continuers/assessments respectively by Schegloff, 1982). Feedback responses have been analyzed through the notions of alignment and affiliation by Stivers (2008) and Stivers et al. (2011):

[…] we conceptualize alignment as the structural level of cooperation and affiliation as the affective level of cooperation (Stivers, 2008). Thus, aligning responses cooperate by facilitating the proposed activity or sequence; accepting the presuppositions and terms of the proposed action or activity; and matching the formal design preference of the turn. By contrast, affiliative responses cooperate at the level of action and affective stance. Thus, affiliative responses are maximally pro-social when they match the prior speaker’s evaluative stance, display empathy and/or cooperate with the preference of the prior action.
(Stivers et al., 2011: 21)

5In line with these authors, Guardiola and Bertrand (2013) showed that alignment can be achieved through generic responses which explicitly mark the construction of shared knowledge while affiliation is preferentially achieved by specific responses that display an evaluative or attitudinal function reflecting the storyteller’s stance. By using specific responses, the listener can also take the other’s perspective leading to a role reversal. Moreover, the authors showed that the ratification by the speaker of the specific response leads to an interactional convergent sequence. In other words, what we have called “interactional convergence” requires not only the aligned and the affiliated response from the recipient but also its ratification – itself aligned and affiliated – by the speaker.

6Like storytelling, listing practice projects a multi-unit turn in which the addressee can only provide generic feedback mh (mm) (see Selting, 2007) whereby he/she reveals a form of alignment showing that for example he/she accepts his/her punctual role as recipient or that he/she indicates a new item to be added to the common ground. However, when the recipient provides more than generic feedback, i.e. a SFR, we argue that he/she orients to the prior utterance by revealing not only a form of alignment but also a form of affiliation, with the recipient exhibiting the same stance as the previous speaker.

7Based on these different observations, the present work focuses on the different interactional trajectories occurring after the SFR within listing practice. We argue that the only way to determine if the collaborative achievement is successful, i.e. leads to a convergent sequence, is to take the complete achievement into account.

8Finally, how a recipient aligns and affiliates by giving appropriate responses to lists that the prior speaker accepts/ratifies or not, requires him/her to take into account the activity type in which lists occur (Lindström & Sorjonen, 2013). Indeed listing practice has been described as a practice or an activity which is always embedded in a larger one (Selting, 2007). In this study, where lists are mainly embedded in storytelling, we suggest that, not only could listing practice be impacted by this larger activity in terms of interactional goals or turn-taking rules, but it could also be impacted by the specific phases of the storytelling. Among various studies on the structure of storytelling, Labov and Waletzky (1966) proposed a formal model of the successive phases, among which the most relevant for our work are: the orientation presenting characters and spatiotemporal information; the complication relating the different successive actions or events leading to the culminating point or apex and the evaluation which expresses a point of view or the implications of the story. Moreover, in work dedicated to a large project such as storytelling in conversation, Selting (2000) added a phase similar to a parenthesis, called “aside”. Considering the time progression of phases, with orientation and complication being mainly produced in the beginning of the story and evaluation at the end (even if they can sometimes occur in different stages of storytelling), it has been shown that generic responses systematically tend to appear at the beginning of the story while specific ones occur at the end (Bavelas et al., 2000; Bertrand & Espesser, 2017). Generic responses are thus likely to be associated with the orientation and complication phases while specific responses are preferentially likely to be associated with the evaluation phase.

9In order to show the strong link between listing practice and its embedding in storytelling, we shall see that ratification by the prior speaker depends on the type of narrative phase in which the list occurs and also on the type of sequence that can be developed within these phases.

3. Corpus and method

10This study was performed on the “Corpus of Interactional Data” (CID) (Bertrand et al., 2008; Blache et al., 2009). The CID is an audio-video recording of French face-to-face conversations (8 pairs of speakers, 8 hours, about 115,000 words). It was recorded in an anechoic room. The participants were asked to talk about either unusual situations (3 dyads) or contentious professional situations (5 dyads) in which they were involved.

11Despite the task and the setting, the protocol was designed to favor conversational interaction. Not only were all the participants colleagues at the same University, but the two members of each dyad were also friends used to meeting each other outside of work. Moreover, bringing together friends for an entire hour allowed for numerous digressions.

12As this corpus was developed in order to provide multimodal annotations at multiple levels, it involved each speaker being equipped with a microphone headset enabling the recording of the two speakers’ voices on different sound tracks in order to allow for a fine-grained analysis of overlap phases, namely at the phonetic and prosodic levels.

13Using Praat (Boersma & Weenink, 2009) the speech signal was pre-segmented into inter-pausal units (IPUs) defined as speech blocks punctuated by at least 200ms silent pause. A manual orthographic transcription was performed within these IPUs. Using this set of IPUs as input and following the same formal annotation scheme, multiple annotations at different linguistic levels (phonemes, syllables, morpho-syntactic categories, discursive units, prosodic phrasing, intonational contours, narratives, disfluencies, gestures) were then performed (Blache et al., 2010).

14Lists were characterized by a syntactic and semantic parallelism and the typical rising list contour or the repetition of the same contour. Thus, our collection is composed of various types of lists including for example the reiteration of a noun phrase, a verbal phrase and so on. A manual annotation based on perceptive identification was performed on the whole corpus by one expert. A second expert checked this annotation. Only consensual cases were retained. Once the lists were identified, only those produced by both partners were examined.

15As said above, the lists always occurred in storytelling. Thanks to the intensive annotation campaign performed on the corpus over recent years, all the annotated narratives (about 150) and within them each formal phase were available.

16Then, we conducted a corpus-based study combining methods from conversation analysis and interactional linguistics (Couper-Kuhlen & Selting, 1996). After identifying lists in a systematic way taking into account syntactic, semantic and prosodic components, we examined SFRs and their evaluation/ratification, step by step, within the structure analyzed in a sequential approach.

4. Structural organization of listing practice

17Listing practice in face-to-face dyadic conversation can be performed by both participants. In this section, we briefly present previous work on this practice, focusing on its relevant features which make list construction a good candidate for illustrating joint activity by both partners. We will illustrate this section with examples extracted from our corpus.

18The necessary starting point for participants in conversation is to identify and recognize what the speaker is currently doing in the conversation. Once this identification has been made, each of them can provide appropriate behavior, i.e. an appropriate response.

4.1. The three-part list

19The preference for list construction as three-part lists or three-part units was demonstrated by Jefferson (1990: 68), who defined a three-part list as “the product of an oriented-to procedure by which lists are properly constructed”. This means that a list constructed with only two items is often completed by what the author called a “generalized list completer” as third item. Examples [1] and [2] illustrate this point.

21Just before the transcript, MB, who was working in an elementary school, was telling AC how shocked she was by some of her pupils’ parents who did not have enough money to live on but who still bought high-tech equipment.

[1]
	1		AC_152	c’est + un peu ça les les les gens qu- même qu’ils sont dans
	2			la misère ils ont quand même
	3	->	AC_153	la télé le satellite le scope
	1		AC_152	that’s + what it’s like, people even when they are
	2			destitute, they still have uh
	3		AC_153	TV, satellite, a DVD player

22Extract [2] shows a list composed of 2 items + generalized list complementer.

23In this excerpt, ML, who had worked in a nursery when she was younger, is reporting a colleague’s behavior with some babies, and in particular, the fact that she let one baby cry alone instead of taking care of him.

[2]
	1	->	ML_143	et puis il était tout triste tout malheureux tout
	2			ça puis je le prends sur moi + je lui fais un gâté a- en
	3			attendant sa maman
	1		ML_143	then he was very upset really sad and everything
	2			then I picked him up + held him and gave him a hug while
	3			waiting for his mother

24The recognizability of a list in progress raises the issue of the number of items involved making a list recognizable as such: Jefferson (1990) considers the first item as sufficient but Lerner (1994) claims that only the second item can retrospectively show that a list is in progress. In the next part, we introduce the prosodic dimension that allows us to corroborate Jefferson’s proposition.

4.2. The prosodic component

25Following Selting (2007), prosody, and more particularly the intonation component, “is used as a resource to methodically make list initiation recognizable for recipients, and further make entire lists interpretable as lists in conversational talk” (Selting, 2007: 488).

26Several intonational contours associated with list items have been identified (see Selting, 2007, for a review on German). Instead of presenting an exhaustive inventory of the different contours involved in lists (which is not the purpose of the present study), here we focus on the crucial role played by intonation as a projection component enabling recipients to recognize and appropriately react to listing practice.

4.2.1. Typical list intonation or melodic cliché

27List prosody in French has been summarized in Di Cristo (2016: 215-216) who describes some continuative intonational patterns as “melodic cliché”. Among them, the rising contour appears as the most prototypical one. Portes et al. (2007) systematically investigated it in the CID. Results showed a significant acoustic difference between the canonical continuative rising contour and the continuative rising list (RL) contour, with a smaller slope for the latter (see Figure 1). The authors interpreted this RL contour as a stylized version of the canonical continuative rising contour both in formal and semantic terms. We would like to highlight here that such a stylized contour will be more easily identifiable to listeners.

4.2.2. Complete/incomplete list

28Couper-Kuhlen (1986) described complete lists as characterized by a successive rising intonation contour ended by a final falling contour or a successive falling contour ended by a rising contour. Conversely, the incomplete list is characterized by only a successive rising or a successive falling contour.

29Extract [3] and Figure 1 show a complete list performed by both participants. IM’s child is left-handed. In the previous minutes, she recounted the way she found out that the teacher forced her son to write with his right hand. In this excerpt, IM, in reported speech, presents the arguments of the teachers to justify her behavior.

[3]
	1	->	IM_728	il faut absolument qu’il s’entraîne de la main droite euh
	2	->	IM_729	il a pris un mauvais pli euh vous l’avez laissé faire enfin
	3	->	ML_691	et moi je vais le redresser quoi
	4		IM_730	han
	1		IM_728	he absolutely must practice with his right hand uh
	2		IM_729	he has got into a bad habit uh you did nothing about it well
	3		ML_691	I I’m going to straighten it out so
	4		IM_730	hm

30Figure 1 shows such a complete list constructed by both partners in French (extracted from Guardiola & Bertrand, 2013). Tiers 1 and 4 present the verbal transcription of the extract, concerning the list initiator and the recipient respectively. Tiers 2 and 5 concern prosodic phrasing: the boundary strength is reflected by an index corresponding to the levels of constituency in French (1 = accentual phrase “ap”; 2 = intermediate phrase “ip”; 3 = intonational phrase “IP”) (Jun & Fougeron, 2000). Tiers 3 and 6 show the intonational contour with regard to the IP level. RL refers to rising list already discussed above (Portes et al., 2007). As a continuative rising contour, RL projects more-to-come, fulfilled by the recipient until the recipient ends the list herself by using a falling contour on the added item and making it the last of the (complete) list. Example [3] is a good illustration of the listing practice as a recognizable and collaborative construction (Selting, 2007: 487, quoting Erickson, 1992).

31It is noteworthy that although a prototypical intonational pattern has been found to characterize lists, not only do other types of patterns exist (see Di Cristo, 2016) but the mere repetition of similar patterns can also contribute to creating a list (Selting, 2007; Szczepek-Reed, 2006).

4.2.3. The virtual list

32As we said above, a stylized contour is more easily identifiable by recipients, even on its first occurrence. Following Portes et al. (2007: 160; our emphasis), on an isolated phrase this contour could trigger “the interpretation of this phrase as the first member of a virtual list; or even on interrupted lists, meaning that the addressee is supposed to finish it mentally”. This is completely in line with the crucial role of prosody also highlighted by Selting (2007: 485) for whom “by using a typical list-intonation for a possible list-item, already first items can be made recognizable as list items”.

33Extract [4] shows such a RL contour in a virtual list.

34This excerpt closes a long narrative in which AG was saying that, in French, we often use the term “marron”, which is a sort of inedible chestnut, instead of using the word “châtaigne” (edible or sweet chestnut), which can be very confusing.

[4]
	1		YM_734	ouais ouais puis en plus c’est ouais la c’est
	2	->		la dinde aux marrons enfin tout + tout est il y a que la farine
	3			de châtaignes quoi
	4	->	AG_684	et ouais tout est au ma- les marrons glacés
	1		YM_734	yeah yeah and besides it’s yeah that is
	2	->		turkey and chestnuts well it’s all it’s all there’s only
	3			chestnut flour you see
	4	->	AG_684	yeah everything is with chest- candied chestnuts

35The two participants relate a story in which a friend of theirs who is from Quebec was misled by the inappropriate use of the term “marron” in French concerning specialities made from “châtaignes” given that only the latter are edible. We can see a single item produced with the typical rising contour (“dinde aux marrons”, line 2) intended to characterize a recipe made with sweet chestnuts despite the term used (“marrons”). While YM continues talking and showing one exception (with “la farine de châtaignes”), AG adds another example of a badly named recipe (“marrons glacés”) produced with the similar rising contour allowing us to infer that he has correctly interpreted the first and single item as a list item.

36In Extract [5] the two participants are discussing the question of choosing which surname a child should take. This example shows another case of a single item produced with a RL contour. In this case, as suggested by Portes et al. (2007), the speaker (AG) could mentally finish this list with an alternative utterance such as “why he takes his”. We can see that although AG (line 5) does not express such an explicit list item, he nevertheless takes it into account as an additional argument that he accepts and from it he elaborates the next idea.

[5]
	1		AG_331	euh de dire tu vois + et puis pour les parents ça peut
	2		YM_371	ouais
	3			être un choc si tu veux de se dire tu vois
	4	->	YM_373	ouais ouais c’est ça pourquoi il a pas pris le mien @
	5		AG_332	ben ouais alors tu vois ça crée peut-être des mè-
	1		AG_331	uh to say you know and then for the parents it can
	2		YM_371	yeah
	3			be quite a shock to think that you see
	4		YM_373	yeah yeah that’s why didn’t he take mine @
	5		AG_332	yeah so you see it can cause some

37The above examples illustrate the prosodic component as a relevant cue not only for the recognizability of list items, even in a single item case, but also for the co-elaboration of lists.

4.3. The three-part structure of a list

38Adopting once again Selting’s approach, listing can be described as an embedded practice: “lists are normally middle parts of a larger three-component structure” (Selting, 2007: 488). Basically, these three components are:

(a) the projection component, projecting more-to-come, i.e. a multi-unit turn to be constructed, either a pre-detailing and/or a general formulation;
(b) the list itself, suggesting the items as part of either a closed or an open number of list items, as a practice of detailing;
(c) the post-detailing component, completing the structure around the list and at the same time tying the list back to the ongoing topic or activity.
(Selting, 2007: 522-523)

39Example [6] illustrates such a three-component structure. Following the conventions of Selting (2007):

● indicates the projection component (PC),
-> indicates list items,
=> indicates the post-detailing component (PDC).

[6]
	1	●	IM_557	non c’est vach(e)ment bien c’est important hein
	2		ML_544	mh
	3	->	IM_558	ils avaient appr- ils faisaient de la cuisine aussi euh
	4	->	IM_559	du jardin euh
	5	->	IM_560	de l’informatique enfin i- de l’anglais
	6	=>	IM_560a	c’était vachement sympa
	7		IM_561	et ils ont fait le même programme en travaillant que
	8			le matin
	9		IM_562	le même programme que tu fais à l’école euh classique
	10		ML_545	mh
	1		IM_557	no it’s good it’s really important uh
	2		ML_544	mm
	3		IM_558	they learn- they did cookery and uh
	4		IM_559	gardened uh
	5		IM_560	computers and some English
	6		IM_560	it was really good
	7		IM_561	they covered the same program by working only
	8			in the morning
	9		IM_562	the same program you do in a regular school
	10		ML_545	mm

40In this example, IM compares a traditional school and a more alternative one in which her son did many activities in addition to the standard academic program. Her PC in line 1 is a positive evaluation of this alternative school while it projects more to come: what is so good and so important? The different list items enable her to explain her point. Her PDC (line 6) is a conclusive formulation very close to the formulation used in the PC (with a reiteration of the same adverb “vach(e)ment”). The recipient only reacts with a simple feedback response just after the PC and another one a little after the PDC (Selting, 2007), once the argumentation seems quite complete. We will develop this point in the next section about appropriate responses.

41To summarize, intonational contours and the PC constitute important devices used by partners to project lists. According to Selting (2007), this three-component structure can be viewed as a holistic entity or a “gestalt” allowing speaker and recipient to produce it together. In the next section, we analyze extracts in which SFRs (or recipient completions) are devices revealing a form of alignment and affiliation in Stivers’ terms (2008) with prior talk (see also Szczepeck-Reed, 2006) after which different interactional trajectories can be deployed.

5. Sequential analyses of some examples

5.1. Evaluation phase of the narrative

42The next examples show listing practice occurring in the evaluation phase of the ongoing narrative. The first four extracts exhibit two different kinds of side sequence. The first one is a word search sequence and the second one is a humorous sequence. The last extract illustrates a collaborative three-part structure.

5.1.1. Side sequence

43As defined by Jefferson (1972: 294), a “side sequence” constitutes a “break” in the ongoing activity: “In the course of some ongoing activity […] there are occurrences one might feel are not ‘part’ of that activity but which appear to be in some sense relevant”. Embedded in a narrative, such side sequences are called “asides” by Selting (2000).

5.1.1.1. Word search sequence

44Extracts [7] and [8] constitute a joint word search activity. According to Goodwin and Goodwin (1986: 52), “searching for a word […] is a visible activity that others can not only recognize but can indeed participate in”.

45The two examples come from the same interaction between AG and YM and concern the following topic: the birth of AG’s baby. YM is a young father and AG and his girlfriend are expecting their first baby. During the two excerpts, the participants speak about the question of being present or not at the birth.

[7]
	L’hystérie / Hysteria
	1		YM_199	moi c’est un peu tout c’était pas tellement le le sang tout ç
	2			enfin ouais si ça m disons que ça me disait rien du tout quoi
	3		YM_201	mais euh mais c’est aussi le
	4	●	YM_202	le côté hystérique un peu de enfin c’est normal tu vois elle
	5			souffre et machin
	6		AG_189	ouais ouais ouais ouais
	7		YM_203	mais l j’ai du j’ai a j’ai du mal avec euh les
	8		YM_204	l’hystérie ou les trucs comme ça et ça fait que c pff je me
	9			suis dit c’est plutôt quelque chose qui va qui va qui va m
	10	->	AG_191	me traumatiser ouais ouais ouais ouais ouais ouais
	11	->	YM_206	ouais me gonfler quoi me euh m’énerver enfin
	12	=>		pas m’énerver mais je sais pas comment dire quoi
	13		YM_207	et du coup euh j bon tu vois si j s si tu prends aucun plaisir
	14			à ça je pense c’est
	15		AG_192	ouais c’est sûr
	1		YM_199	For me it was kind of the whole thing, not really the the blood
	2			all of th- well yeah let’s just say that I didn’t, like the idea
	3		YM_201	but uh but it’s also the
	4		YM_202	the hysterical side kind of well it’s normal you know she’s
	5			in pain and all that
	6		AG_189	yeah yeah yeah yeah
	7		YM_203	but th- I have some I have to I’m not too good with uh the
	8			hysteria and all that and so - I said to myself it was it would
	9			be something that would that would that would
	10		AG_191	traumatize me yeah yeah yeah yeah yeah yeah
	11		YM_206	yeah well upset me bother me well not bother but I don’t
	12			know how to explain
	13		YM_207	and so uh I well you know if I if you don’t get any pleasure
	14			from it I think it’s
	15		AG_192	yeah definitely

46YM, a young father at the moment of the recording, tries to explain why he would have preferred not to be present at his baby’s delivery if it had been possible. But he has difficulty explaining his feelings, which can be seen through various discursive cues: mitigations (“un peu tout, tout ça”, line 1), vague lexicalization (“machin”, line 5), and many disfluencies (filled pauses). AG wants to help YM and proposes a verb which could correspond to the feelings YM is trying to express. By suggesting a word, in addition included in self-reported speech (“me traumatiser”, line 10) allowing AG to put himself in YM’s place (see Guardiola & Bertrand, 2013), AG shows his orientation toward the current activity and the main speaker’s stance. In doing so, AG aligns and affiliates. YM does not ignore his suggestion but rather takes this word into account (“ouais”, line 11) and makes this proposition the first item of the upcoming list. Then, he proposes a second item which mitigates the first one. Possibly, YM’s ratification while introducing a less dramatic word is not just a matter of facework (Goffman, 1967). Even if the verb “traumatiser” is too dramatic, it seems to help YM who proposes another one (“me gonfler”, line 11), and then, a third one (“m’énerver”, line 11). This example is in line with what Lerner (1994) said about the “response list” as a means for neither accepting nor rejecting a previous proposition.

47Since the function of this list is unusual (a word search), its PC is quite vague. Indeed, “hystérique” does not project, per se, a list. And no list would have been expected if the word had been the right one. However, “hystérique” is the object of the numerous hesitations YM produces in order to mitigate it. And these hesitations disrupt his ongoing discourse. In other words, YM’s entire contribution (lines 4-9) could be considered the PC: “hystérique” which leads to some hesitations, which in turn leads to the word search sequence.

48Just after the third item, YM produces a rather long turn (lines 11, 12) which could be considered the PDC. Giving up on finding the right word (“je sais pas comment dire quoi”, line 12), he closes the list with AG’s agreement who produces various positive feedback signals to punctuate YM’s discourse.

49Taking into account both the nature of the various elements described above and their goal, this list creates a side sequence in which the interactional convergence is particular. Indeed, while the two participants converge to develop a word search sequence, they ultimately fail to find the right word.

50Extract [8] is also a word search sequence. However in contrast to [7], the main speaker, AG, finds the word he was looking for.

[8]
	Tourner de l’œil / Passing out
	1		AG_164	moi j’ai
	2		AG_165	*
	3		YM_182	toi ouais ouais
	4		AG_166	moi de moi moi je me suis enfin si tu veux je
	5			normalement je de enfin
	6		AG_167	si tout va bien je vais essayer de le faire
	7		YM_183	ouais
	8		AG_168	mais euh
	9		YM_184	ouais
	10	●	AG_169	mais j’en sais rien si euh si tu veux je vais pas
	11	->		tourner de l’œil ou j’en sais rien ouais tu vois voilà
	12	->	YM_186	ouais et le résultat ouais ce que tu vas garder
	13			comme image ou euh
	14		AG_171	ouais voilà
	15		YM_187	mais euh
	16	=>	AG_172	parce que
	17		AG_173	j’ai vu tu sais on a prép on a fait les cours de préparation à
	18			l’accouchement là
	1		AG_164	I I
	2		AG_165	*
	3		YM_182	you yeah yeah
	4		AG_166	I for me I I have well
	5			if you like
	6		AG_167	I if all goes well I well I’m going to try and be there
	7		YM_183	yeah
	8		AG_168	but uh
	9		YM_184	yeah
	10		AG_169	but I don’t know whether uh if you like I’ll pass
	11			out or something yeah you see there
	12		YM_186	yeah and in the end yeah what sort of picture it’ll leave
	13			in your mind you uh
	14		AG_171	yeah that’s it
	15		YM_187	but uh
	16		AG_172	because
	17		AG_173	I went to see you know we prep- we went to prenatal
	18			classes

51At the time of the recording, AG’s girlfriend is pregnant. AG is producing a long narrative saying that he would like to be present at his baby’s delivery but, at the time, he doesn’t know if he will be able to cope with the situation (line 10). This last utterance (“mais j’en sais rien”) functions as the PC introducing a list in which all the items are potential feelings. The discursive clues of this word-search sequence are a disfluency (“euh”) with a strong lengthening followed by an unusually long silent pause (which allows the recipient the opportunity to speak and to produce more items) and a mitigation (“si tu veux”) which appears as an anticipated concession for an expression he knows to be approximate (“tourner de l’œil”).

52What is remarkable in [8] is the fact that the first two items of the list are produced quite simultaneously (in overlap) by the two participants. This confirms the relevant projective dimension of the PC. Both speakers used a typical RL contour when producing each item: AG’s proposition – which can be considered a virtual list constructed by a single item – becomes a co-constructed list with the two items added by the recipient. Moreover, the overlap, far from showing a conflict between participants, rather indicates a very collaborative sequence.

53By reiterating his acknowledgement feedback (“ouais voilà”, AG, line 14), the main speaker shows his orientation to the specific feedback from the recipient that illustrates a convergent sequence.

54Finally, AG closes the list by beginning another storytelling (lines 16-18). These last utterances function as the PDC because they may turn out to be the reason and the trigger of his own apprehension concerning his baby’s delivery.

5.1.1.2. Humor and virtual list

55Extract [9], still between AG and YM, shows the end of the story presented in Examples [1] and [2] and constitutes an evaluation phase of the narrative. AG finishes telling the night when his girlfriend gave birth.

[9]
	Elle est occupée / She’s Busy
	1		YM_263	ah ouais nous on est rentré à
	2		AG_236	mh
	3		YM_264	dix heures dix heures et demi je crois du soir
	4		YM_265	et elle a accouché à six heures je crois
	5		YM_268	donc c’était ouais c’était quand même assez long quoi
	6		AG_237	ah quand même ouais
	7		YM_270	parce que ouais c’était euh
	8		AG_239	en plus de nuit quoi
	9		YM_272	ouais ouais putain @ quatre heures moi j’étais là j’en
	10		AG_240	et eh
	11		AG_241	pouvais plus quoi
	12		AG_242	et ouais c’est sûr ouais
	13	●	YM_273	je suis allé dormir un peu dans m enfin bon par rapport à
	14	->		elle c’est sûr que bon @ mais bon elle elle est occupée
	15		AG_243	ouais ouais non mais *
	16			elle a toi tu es là tu @
	17	->	AG_244	@ elle a un but ouais
	18	->	YM_274	ouais elle a euh elle fait quelque chose quoi toi tu es
	19	=>		là tu @
	20		AG_245	ouais
	21		YM_276	non je déconne mais euh
	22		YM_277	mais c’est ouais c’est assez long quoi euh il y en
	23			avait marre quoi
	24		AG_246	ouais
	1		YM_263	oh yeah we went in at
	2		AG_236	mm
	3		YM_264	ten ten thirty pm roughly
	4		YM_265	and she gave birth at six am I think
	5		YM_268	so it was yeah it was quite long in the end well
	6		AG_237	ah right yeah
	7		YM_270	because yeah it was uh
	8		AG_239	and in the night as well
	9		YM_272	yeah yeah god @ four o’clock I I was there I
	10		AG_240	eh eh
	11		YM_272	couldn’t take anymore well
	12		AG_242	yeah sure yeah
	13		YM_273	I went to sleep for a while, well compared to her
	14			of course that well @ but well she was she was busy
	15		AG_243	yeah yeah but still
	16		YM_273	she had you you are there you @
	17		AG_244	@ she had a goal yeah
	18		YM_274	yeah she had uh she was doing something well you you are
	19			there you @
	20		AG_245	yeah
	21		YM_276	no I’m kidding but uh
	22		YM_277	but it’s yeah it was quite long well uh I had
	23			had enough well
	24		AG_246	yeah

56YM is producing a long narrative sequence, recounting the circumstances of his child’s birth. He and his pregnant girlfriend arrived at 10pm at the maternity hospital and waited all night for the child to be born. Insisting heavily on the length of the labor (lines 5-9), YM produces a kind of complaint sequence or, at least, he seems to have this impression (“enfin bon par rapport à elle”). In order to counterbalance the impression he has (and maybe he gives to AG), he chooses to switch to humor to continue telling his story. In other words, in case he has complained inappropriately, and considering that the best form of defense is attack, he exaggerates his complaint, playing a man who is to be pitied more than the future mother because, in this precise case, “she was busy” (“elle est occupée”, line 14). The item proposed by YM initiates the humor, not so much to develop on the way he spent that particular night but, most likely to make fun of himself and the way he had begun presenting the story. “Elle est occupée” (line 14) is an item of a virtual list (with the typical RL contour) to which AG orients with the SFR (“elle a un but”, line 17). Semantically and prosodically aligned with “elle est occupée”, AG’s utterance appears as a second item of the virtual list. Playing along (Attardo, 2002; Hay, 2001) with YM’s humor, AG not only aligns but also affiliates. The feedback he produces after his item (“ouais”, line 17) highlights this affiliation because it could be paraphrased by “I know what you mean”.

57AG, overlapping with YM, and laughing, builds on YM’s humor (“elle a un but”). YM ratifies AG’s intervention by producing a third item (“elle fait quelque chose quoi”, line 18) still in the same vein. This list, co-constructed by the two participants is thus highly convergent because each item is accepted by the other and builds upon the previous one, which is a typical pattern of co-construction of a humorous sequence (Priego-Valverde, 2006).

58Finally, YM produces the PDC (“non je déconne mais euh”, line 21) which also closes the humorous sequence and he returns to the serious mode (Skalicky et al., 2015) he initiated, which is once again accepted by the recipient (AG “ouais”, line 24).

59In sum, this excerpt is remarkable for two reasons. Firstly, because the specific function of the list is to create a co-constructed humorous sequence by the two participants. Secondly, because the list initiated by YM, the main speaker, is in the beginning, just a “virtual” list, which only becomes more real thanks to the listener’s behavior (AG).

60Extract [10] shows another humorous side sequence embedded in the evaluative phase of a narrative. LJ is telling an anecdote about archeological digs he made when he was younger and more precisely, about some people he met there.

[10]
	I.R.A.
	1		LJ_320	euh son père c’est lui qui diri- qui est architecte euh qui
	2			travaille à
	3	●	LJ_321	à l’I.R.A.A. euh Institut de Recherche sur l’Architecture en
	4		AP_401	à l’I.R.A., ah
	5	->	AP_402	il travaille à l’I.R.A.
	6		LJ_321a	@
	7	->	LJ_322	et il plastique euh
	8	->	AP_403	m- ah @ et il pose euh les détonateurs euh
	9	->	LJ_323	ouai- il est horloger à l’I.R.A. @
	10	=>	AP_404	ah c’est pas drôle ça c’est pas
	11		LJ_323a	non c’est I.R.A.A. A. euh Archite- euh Institut de
	12			recherche s- sur l’Architecture Antique
	13		AP_405	I.R.A. ah
	14		AP_405a	oh putain
	1		LJ_320	and his father is the man who mana- who is an architect uh who
	2			works at
	3		LJ_321	at I.R.A.A. uh the Architectural Research Institute on
	4		AP_401	at I.R.A. ah
	5		AP_402	he works at I.R.A.
	6		LJ_321a	@
	7		LJ_322	and he blows up uh
	8		AP_403	ah @ and he puts the detonators uh
	9		LJ_323	yeah he is watchmaker at I.R.A. @
	10		AP_404	ah that’s not funny
	11		LJ_323a	no it’s I.R.A.A. uh Archit uh Research Institute on
	12			antique architecture
	13		AP_405	I.R.A.
	14		AP_405a	Oh God.

61In this sequence, the two participants are speaking about an archaeological dig LJ did some years ago. This excerpt is part of a long narrative sequence about this topic that includes the telling of various anecdotes. LJ evokes one person who was working at the IRAA, i.e. the Research Institute of Antique Archaeology. In a serious frame of mind, LJ begins to quote the name of the institute, spelling out all the different letters of the acronym. At that point, there is nothing to announce the projection of a list, neither a potential prosodic clue, nor any discursive material: LJ has not finished his spelling and no PC can be considered as such.

62Despite the fact that LJ clearly pronounces the two A (the corpus shows it), AP, overlapping and interrupting LJ, rebounds on this acronym to voluntarily evoke the Irish terrorist organization (IRA), whose acronym is phonetically very close. Thanks to a script opposition (Raskin, 1985; Attardo & Raskin, 1991) between architecture and terrorism, AP switches into a humorous mode of communication which is prosodically framed (Bertrand & Priego-Valverde, 2011). Producing this humorous utterance, AP disrupts the ongoing talk (Norrick, 1993) and initiates a side sequence whose main goal is the production of humor. This goal is totally accepted by LJ who begins to laugh (line 6) and immediately adds (“il plastique”, line 7) exhibiting a prosodic orientation (Sczcepek-Reed, 2006). In doing so, he plays along with AP’s humor and participates in the construction of a real humorous sequence. In this regard, AP’s utterance (“à l’I.R.A.”, line 9) constitutes both the trigger of his humorous sequence and the PC which is developed by the first item (“il travaille à l’I.R.A.”, line 5). LJ plays along with AP’s humor by adding a second item (“il plastique”, line 7). It is worth noting here that the two items are produced with a similar intonational configuration. Laughing, AP accepts and plays along with LJ’s humorous utterance and produces the third item (“il pose les détonateurs”, line 8). LJ does the same: after a positive feedback (“ouais”), he adds the fourth item (“il est horloger à l’I.R.A.”, line 9) and also laughs.

63The humorous co-constructed list is highly convergent: firstly, both participants align, adding elements totally oriented with the previous utterance. Secondly, each item is necessarily ratified by the other because it takes the humor even further. Thirdly, this humorous side sequence is initiated by the recipient and not by the main speaker. Last but not least, this co-built humorous sequence disregards the canonical roles of a narrative sequence: each of them is a main speaker.

64Finally, AP, the instigator of the humorous sequence, produces the PDC which closes the sequence by returning to a serious mode, invalidating the humor (“c’est pas drôle”, line 10). Once again, LJ accepts this reversal of position, repeating, this time in full, the meaning of the acronym (line 11) as if no humor had been produced. The parenthesis is closed.

65In sum, the last two extracts show listing practice as used to introduce a new frame, i.e. humor which disrupts the ongoing topic. Whether the first list item is produced by the main speaker or the recipient, it can be the trigger for changing the initial interactional trajectory, on which partners highly converge.

5.1.2. Co-constructing the three-part structure

66Extract [11] occurs in an evaluation phase at the end of the narrative sequence. MB, who is working in a school, is talking about parents who do not have money enough to feed their children properly but who buy high technology devices. Thus, the participants are involved in a sequence while they evaluate parents’ behavior.

[11]
	Les télés et les DVD / TVs and DVDs
	1		AC_150	putain @
	2	●	MB_142	mais s’il y avait que ça qui comptait tu vois avoir
	3	->		douze télés des DVD des godasses trucs enfin c’est voilà
	4		AC_151	ouais ouais
	5		AC_152	c’est + un peu ça les les les gens qu- même qu’ils sont dans la
	6			misère ils ont quand même
	7	->	AC_153	la télé le satellite le scope
	8	->	AC_154	à côté de ça ils bouffent rien quoi mais ils ont un super
	9	->		équipement machin euh
	10		MB_143	mh
	11		AC_155	les gamins i(l)s sont habillés à l’arrache
	12		MB_144	ouais et encore
	13	=>	AC_156	c’est souvent comme ça maintenant c’est de plus
	14			plus en plus comme ça ouais
	1		AC_150	God. @
	2		MB_142	as if that was the only thing that mattered you know having
	3			twelve TVs DVDs shoes and stuff well there you go
	4		AC_151	yeah yeah
	5		AC_152	that’s + what it’s like people even when they are destitute they
	6			still have
	7		AC_153	TV satellite a DVD player
	8		AC_154	at the same time they eat nothing but they have great
	9			gear uh
	10		MB_143	mm
	11		AC_155	their kids are badly dressed
	12		MB_144	yeah and worse
	13		AC_156	it’s often like that now it’s more and more
	14			like that yeah

67MB, the main speaker, produces a PC (“s’il y avait que ça qui comptait”, line 1) which is immediately followed by three items clarifying which is underlined by “ça” (“douze télés des DVD des godasses”, lines 3-4). She ends her turn with two discourse markers (“enfin c’est voilà”), which could be considered the PDC, under their apparently conclusive function. In this case, the numerous items produced by AC (“la télé le satellite le scope à côté de ça ils bouffent rien quoi mais ils ont un super équipement machin”, lines 7-9) would be produced too late to be taken into account and would constitute a second different list, still semantically related to the first one but not inserted in it. Nevertheless, MB takes into consideration the items AC adds, firstly with a minimal feedback (“mh”, line 10) and then more explicitly (“ouais et encore”, line 12).

68Another interpretation of why the items are added by the recipient, accepted and considered to be legitimate elements of the list by the main speaker (which thus becomes co-built) could be hypothesized: it is possible that the conclusive discourse markers MB produces (“enfin c’est voilà”, line 3) are not really discourse markers. In French conversation, these kinds of markers can be conclusive but they are mainly produced routinely allowing the speaker to indicate that he/she has finished not because he/she has produced sufficient elements to close his/her demonstration – which would render whatever element added by the recipient parasitic – but because he/she cannot think of anything more to add. In the latter case, the door remains open for elements the recipient may add to be relevant.

69Considering this explanation, this sequence is highly convergent: the recipient is given the opportunity to speak. By producing the same listing device (alignment) and by sharing the same values (affiliation) the recipient’s response can easily be ratified by the main speaker. Moreover, at a structural level, the whole three-component structure of the list is initiated by MB, the main speaker, and closed by AC, the recipient.

5.2. Orientation phase of the narrative

70The orientation phase presents characters and spatiotemporal information and is usually produced in the beginning of the story.

5.2.1. Minimal taking into account

71Extract [12] is the only case that is structurally outside the narrative although it is linked to the narrative, since the main speaker introduces new elements which could be considered elements of an orientation phase in a usual narrative. In this excerpt, YM tells the way he was welcomed as a post-doctoral student, at the University of Edinburgh.

[12]
	La carte / The Card
	1		YM_1245	mais c’est vrai en même temps que quand tu es euh du coup
	2			quand quand tu arrives je veux dire tu as aussi euh
	3		AG_1106	ah voilà
	4	●	YM_1246	tu as tu vois tu as tout quoi je veux dire on te file
	5	->		le bureau la clé euh le passe pour rentrer dans la fac
	6	=>		euh tout quoi
	7		AG_1107	ah ouais toi tu
	8		AG_1108	ouais c’est sûr c’est sûr c’est c’est vrai c’est vrai
	9		AG_1111	ouais
	10	->	AG_1112	carte étu carte de bibliothèque euh ouais
	11		YM_1249	donc
	12		YM_1250	ouais tu as tu as donc tu arrives tu es tu es vraiment voila tu
	13			as euh
	14		YM_1251	et puis tu es tu es euh
	15		YM_1252	tout le mon on te présen- enfin tu vois c’est vraiment voilà tu
	16			arrives tu es bon pas à tout le monde parce que c’est c’est
	17			grand quoi
	18		AG_1118	mhm mh
	1		YM_1245	but it’s true at the same time that when you are uh so when
	2			when you arrive I mean you have also uh
	3		AG_1106	oh well
	4		YM_1246	you have you know you have everything well I mean
	5			they give you a study a key uh the passcard to get into the
	6			university uh everything
	7		AG_1107	oh yeah you
	8		AG_1108	yeah definitely definitely it’s it’s true true
	9		AG_1111	yeah
	10		AG_1112	student card library card uh yeah
	11		YM_1249	so
	12		YM_1250	yeah you have you have so you arrive you are you are really well
	13			you have uh
	14		YM_1251	and then you are you are uh
	15		YM_1252	everybo- they introdu- well you see it’s really
	16			well you arrive you are well not to everybody because it’s it’s
	17			huge
	18		AG_1118	mm mm

72This excerpt is very interesting because, if at first glance, it shows a long co-built list (5 items) a more careful observation reveals the importance of timing which manages the recipient’s specific response. Indeed, it is not enough to add an element to the ongoing list (even if it is semantically connected with the previous items); this item has to be added at the right time to be taken into account, which is not really the case here.

73In the narrative sequence about the university staff preceding this extract, YM, the main speaker, produces an entire three-part list in a very canonical form. He begins by the PC (“tu as tout quoi”, line 4), immediately followed by the three items (“le bureau la clé euh le passe”, line 5) of the list which develop the PC. He closes the list with the PDC (“tout quoi”, line 7) which summarizes the previous items.

74After YM’s PDC, AG produces two more items (“carte étu carte de bibliothèque euh”, line 10) which are apparently acknowledged by the main speaker, as a concession (“ouais”, line 12), before immediately repeating the contextualization elements (“donc tu arrives”, line 12, already said in line 2). We consider that the reasons why AG’s turn is just acknowledged and not really ratified by YM are twofold. The first one is probably a matter of face-work. The second, however, is a matter of timing. On the one hand, as we said above, this extract can be seen as a kind of orientation phase in which the recipient could be less likely to intervene with specific feedback, unlike the evaluation phase in which participants can share speech more symmetrically (see previous section). On the other hand, considering the canonical structure of lists, the fact that the main speaker has already produced the PDC means that this contribution is mistimed.

5.2.2. Refusal of the recipient’s item

75Extracts [13] and [14] concern the same interaction between AC and MB. MB, who is working in a school at the same time as pursuing her studies, tells a long anecdote about some of her colleagues. The two excerpts occur in the narrative orientation phase.

[13]
	Tu fais que tchatcher / You Just Talk
	1		MB_316	donc ça c’était je sais plus y a deux ans ou autre
	2		MB_317	y a une nana qui est arrivée dans l’école
	3		MB_318	et ce que je supportais plus c’était les
	4	●	MB_319	c’est qu’en fait comme ça se passe à midi
	5		MB_320	tout le monde c- d- commence à arriver avec de la bouffe et
	6	->		des bouteilles et tu bois tu bouffes et tout et en fait de
	7	=> ●		réunion tu fais rien du tout tu fais que tu fais que bouffer
	8	->	AC_403	ouais tu fais que dalle tu fais tu tchatches quoi *
	9	->	MB_321	dire des conneries tu avances pas
	10		MB_322	et euh je disais moi si je veux bouffer avec des gens je choisis
	11			j’ai je mange avec qui je veux avec mes copains mais pas avec
	12			euh tronche machin euh sous prétexte de
	13		AC_404	ah ouais
	1		MB_136	so it was I don’t remember two years or so ago
	2		MB_317	there was a girl who arrived at school
	3		MB_318	and what I couldn’t stand was the
	4			was how things happened at lunchtime
	5		MB_320	everybody starts coming in with food and bottles and you drink
	6			you eat and everything and instead of having a meeting you
	7			don’t do anything the only thing you do you just eat
	8		AC_403	yeah you do nada you do you talk
	9		MB_321	talk shit and don’t get anything done
	10		MB_322	and uh I was thinking if I want to eat with people I choose I
	11			I eat with who I want with my friends but not uh thing-face
	12			like on the pretext
	13		AC_404	oh yeah

76In this sequence, MB tells a story concerning her work and the bad habit (according to her) her colleagues have of scheduling meetings at lunchtime. She explains that such a schedule is totally counterproductive and she initiates a list in which each item is proof of this counterproductivity. The list is as follows.

77MB, the main speaker, presents the setting of her story: how things happen at lunchtime (“comme ça se passe à midi”, line 4). It is also the PC of her list because it allows her to propose all the various activities it is possible to do at noon, despite the fact that all the characters of her story are in a meeting. She thus proposes three items of her list, which are both the activities her colleagues do but also the list of the grievances she has about them (“tout le monde commence à arriver avec de la bouffe et des bouteilles et tu bois et bouffes et tout ça”, lines 5-6). Following these items, she also produces the PDC which functions both as the summary of the various items (“et en fait de réunion tu fais rien du tout”, lines 6-7) and as a new PC followed by a first item (“tu fais que bouffer”, line 7) initiating another list.

78Overlapping MB’s second list, AC proposes two more items (“tu fais que dalle tu tchatches”, line 8) apparently aligned and affiliated insofar as they show an adaptation to the activity (list item) and to the main speaker’s stance (sharing the same kind of grievance). Despite this fact, the recipient’s response is pragmatically irrelevant for two reasons. Firstly, these items could be considered redundant by MB because they are a sort of summary of what MB had just said. Secondly, these items could be considered parasitic because they are produced in overlap while the main speaker wants to go on talking (high intensity and pitch), as the remainder of the example shows. Indeed, MB, also in overlap, continues her list with a last item (“dire des conneries”, line 9) and the PDC (“tu avances pas”, line 9).

79The rude manner in which MB, the main speaker, shows that she wants to continue speaking (not taking into account AC’s intervention and keeping on talking to produce her two lists successively), can be explained by the fact that MB is engaged in the orientation phase of her story. Indeed, in lines 10-12, MB finally expresses her own conception of lunchtime saying that she wants to choose the people with whom she has lunch. In other words, the goal of the two lists she produced just previously is not to illustrate something she had already said, which could have allowed the listener to co-elaborate the discourse, at the very least to show her agreement, but to present the characters and circumstances of events in the story and then prepare the ground for establishing her own opinion and justify it in advance. In this regard, whatever AC says could only be refused, at least based on timing.

80Extract [14], has the same design as [13].

[14]
	La femme de service / The Cleaning Lady
	1		MB_362	je bois pas à midi quand j’ai classe l’après-midi
	2			ou je sais pas quoi tu vois en plus eh ouais
	3		AC_436	ben non hein c’est c(l)air
	4		AC_437	ouais ouais
	5		MB_363	hurler comme des hystériques
	6		MB_364	et à ce moment là arrive une femme de service qui venait euh
	7	●	MB_365	pour je sais pas quelle raison et y avait tellement de bordel
	8	->		dans cette salle entre les autres qui regardaient les photos en
	9	->		hurlant de rire les autres qui picolaient et tout
	10	->	AC_438	et vous qui vous prenez la tête
	11	=>	MB_366	euh l’autre qui continuait à ronchonner alors qui faisait
	12		MB_367	oui mais moi des gens comme ça tu comprends gnagnana puis
	13			elle allait rincer les trucs à l’évier elle parlait en me tournant le
	14			dos mais elle me parlait à moi et moi qui disait si tu veux
	15			me parler regarde-moi en face au lieu de parler au mur @ un truc
	16			de fou
	17		AC_439	mh
	18		AC_440	@
	1		MB_362	I don’t drink at lunchtime when I’m teaching in the afternoon
	2			or whatever you know as well
	3		AC_436	well no uh of course not
	4		AC_437	yeah yeah
	5		MB_363	screaming as if they were hysterical
	6		MB_364	and then a cleaning lady arrived who was coming in for uh
	7		MB_365	I don’t know why and there was such a mess in the
	8			room between them looking at the photos and screaming with
	9			laughter and the others having a drink and all
	10		AC_438	and you getting all worked up
	11		MB_366	uh she kept on complaining going like
	12		MB_367	yeah but I people like that you know blah blah blah
	13			then she went to wash some things in the sink she was talking
	14			with her back to me but she was talking to me and I was saying
	15			if you want to talk to me look me in the face instead of talking
	16			to the wall @ it was crazy
	17		AC_439	mm
	18		AC_440	@

81MB produces a list with a PC (“tellement de bordel”, line 7) immediately followed by two items in order to make explicit what she calls “bordel” (“les autres qui regardaient les photos en hurlant de rire les autres qui picolaient”, lines 8-9).

82At that point, and without overlapping MB’s turn (unlike Example [13]), AC produces a third item (“et vous qui vous prenez la tête”, line 10). Her item is not only aligned and affiliated but also produced inside the structure of the list: she adds the item before MB produces the PDC. The list is thus not finished. However, MB does not take AC’s item into account at all and keeps her own discourse going, producing the PDC (“et l’autre qui continuait à ronchonner”, line 11) as if AC had never participated.

83Once again, the fact that MB’s behavior does not take into account AC’s efforts to co-elaborate the discourse is very questionable. The reason can be found in MB’s discourse immediately following her list. Indeed, it turns out that the real topic of her story is not what a “mess” (“bordel”) work meetings can be but more specifically the behavior of one of her colleagues. In other words, the first two items of the list are there to stage the setting and to highlight her colleague’s behavior, considered inappropriate by MB and which she thinks is wrong. The focus on the real topic of her story can be seen by the reported speech and her long intervention following (line 12).

84In sum, even if recipients’ responses are aligned and affiliate with the main speaker’s discourse and are produced at what appears to be the right moment (before the end of the list and before the production of the PDC), considering the type of activity in which lists are embedded (a long narrative), the list initiator considers him/herself the main speaker and does not want to relinquish his/her turn before having addressed the real topic of his/her story. In other words, specific responses can be considered illegitimate, in terms of participation rules. These observations corroborate those reported by Stivers (2008: 36) showing that such a specific response arriving too early – for example because the story is considered incomplete by the main speaker – means a misinterpretation of what really happens and then becomes a disaligned response.

5.3. Complication phase of the narrative: stopping a list

85Just after the orientation phase, the complication phase concerns the different successive actions or events leading to the culminating point (apex) of the narrative.

86Extract [15] shows a case in which the recipient’s response conveys a particular illocutionary force, despite its relevance in terms of alignment and affiliation. This excerpt contains the complication phase. IM tells an anecdote about her son’s teacher who wanted to force him to write with his right hand whereas he is left-handed.

[15]
	Le gaucher / The Left-Handed Child
	1		IM_726	je cours à l’école + je lui dis
	2		IM_727	vous savez que ça fait cinquante ans qu’on laisse les enfants
	3			écrire avec la main qu’ils veulent hein
	4		ML_690	@
	5	●	IM_728	oui oui mais vous comprenez il aura une horrible je ne peux
	6	->		pas tolérer euh une chose pareille euh il faut absolument qu’il
	7	->		s’entraîne de la main droite euh
	8	->	IM_729	il a pris un mauvais pli euh vous l’avez laissé faire enfin
	9	->	ML_691	et moi je vais le redresser quoi
	10	=>	IM_730	han
	11		IM_731	oh là là je dis bon ben écoutez dans ce cas nous n’avons plus @@
	12			rien à nous dire @@ je le change d’école immédiat(e)ment quoi je
	13			l’ai encore changé d’école
	14		IM_732	et là
	15		IM_733	Dieu merci je suis tombée sur une maîtresse
	16		IM_734	enfin bien
	17		IM_735	parce que celle d’avant l’avait enfermé dans le placard + à
	18			balai
	19		ML_692	@
	1		IM_727	I rush to the school and say to her
	2		IM_727	do you know that we’ve been letting children write
	3			with whichever hand they want to write with for over fifty years
	4		ML_690	@
	5		IM_728	yes yes but you do understand he will have horrible I
	6			couldn’t tolerate uh such a thing uh he absolutely must practice
	7			with his right hand uh
	8		IM_729	he has got into a bad habit uh you did nothing about it
	9		ML_691	and I I’m going to straighten it out so
	10		IM_730	hum
	11		IM_731	Oh boy I say well listen in that case we don’t have @@ anything
	12			else to say to each other @@ He is changing schools
	13			immediately so he changed schools again
	14		IM_732	and there
	15		IM_733	thank god I’ve at last found a teacher
	16		IM_734	a good one
	17		IM_735	because the previous one had him locked in the broom +
	18			closet
	19		ML_692	@

87IM, the main speaker, is producing a very long narrative concerning the behavior of her son’s teacher when he was a little boy. She explains that by chance she realized that the teacher was forcing her son to write with his right hand even though he was left-handed. Considering this totally unacceptable, IM, in a segment of reported speech introduced by the PC (“vous comprenez”, line 5), lists five items which are the arguments the teacher gave to justify herself (lines 5-7).

88If IM’s list is canonical on a prosodic level, its structure (five items) is unusual. Immediately after the last item, the recipient (ML) produces a sixth item (line 9, with a final falling contour) that completes the list (see Figure 1). Thus, both from a discursive and prosodic level, ML aligns and affiliates, summarizing the teacher’s behavior presented by the main speaker, which is highlighted by the conclusive discourse marker “quoi”. But, considering that three items are the preferred structure of a list, this sixth list item could, on the contrary, signify to IM that ML has understood and that this phase in IM’s story was probably too long. It is probably no coincidence that the canonical structure of a (co)list is constructed with three items. This number is the right balance: fewer items could make it difficult for the recipient to recognize a list structure; too many could complicate the smooth progress of the interaction itself because they could appear redundant and perhaps parasitic. Moreover, as the list items are produced in the complication phase, if they are too numerous, they can delay the appearance of the apex and trigger a kind of impatience in the recipient. Therefore this remarkable example shows that while the main speaker can usually claim the right not only to speak, but also to consider the recipient’s contribution legitimate or not, in a narrative sequence, the recipient also has his/her say about the discourse.

89Finally, by choosing to complete the list, the recipient ML won the power balance because IM cannot develop her list any further and continues her story with the reaction she had in response to the teacher.

90In other words, at first glance, the co-construction of this list seems to be convergent but simultaneously, its closure, initiated by the recipient, can be considered very intrusive. This then makes it difficult to consider such a sequence truly convergent.

6. Concluding remarks

91This work is part of a larger project aiming at better characterizing interactional convergence in French conversation, which requires alignment and affiliation in Stivers’ terms (2008) from both speaker and recipient.

92In this study, listing practice has been observed as an activity collaboratively achieved by two participants. The aim of this article was twofold: first, to contribute to the sequential environment study of listing practice in conversation, and second, to treat it as a gateway to investigating the larger issue of interactional convergence.

93As an observable per se, the analysis confirms the three-part structure of a list (PC, items of the list, PDC), previously identified by Selting (2007). More importantly, the analysis of several extracts has shown that this preferred structure is not random.

94The examination of interactional trajectories following a SFR in collaborative listing practice shed light on the ratification by the main speaker that numerous studies have neglected in collaborative sequences. We have demonstrated that this ratification depends both on the context within the larger activity in which it is embedded (i.e. storytelling and its different phases), and the types of sequences (e.g. side sequence) that the ratification contributes to creating. Then when specific feedback occurs in a narrative evaluation phase and/or with the right timing within the list structure (before the end of the three-part structure), it is more likely to be followed by the main speaker’s ratification, making the sequence a convergent one. When specific feedback occurs in a narrative orientation phase and/or is mistimed within the list structure (i.e. too early or too late such as after the PDC), it is more likely to be followed by a turn that does not promote the sequence as convergent one.

95More generally, this study confirms the proactive nature of feedback (Tolins & Fox Tree, 2014). It supports the idea that the same specific feedback, which at first glance reflects the features of alignment and affiliation that make the collaborative listing practice a relevant candidate for convergence, can also become an inappropriate response (misaligned response) resulting in a less interactional convergent trajectory. So, behind the apparent collaboration – until now only based on the feedback response from the recipient – one cannot know if the interactional convergence is successfully achieved without taking into account the third element of ratification within listing practice.

Transcription conventions

• Elision: the characters related to the omitted phonemes are written between parentheses.
  “petit” pronounced /pti/ is noted p(e)tit.

• Truncated words: annotated with a final dash.
  le li- le livre (the book).

• Missing liaisons: #
  trois # amis (missing required liaison).

• Onomatopoeia: the typical back-channel onomatopoeia /m/ produced by the hearer is transcribed as “mh” when it was realized with one syllable, and “mhm” for two syllables.

• Incomprehensible sequences annotated with a star: *

• Laughter: @
  Said while laughing: @@ …@@

• Pauses: long pauses (more than 200ms) are automatically detected and enable to identify IPU.
  The shorter perceptible pauses are notated with “+”.

• Overlaps: overlapping words or utterances are underlined.
  ouai- il est horloger à l’I.R.A. @
  ah c’est pas drôle ça c’est pas