Joint planning in conversations with a person with aphasia

This study explores practices employed by a person with aphasia (PWA) and his wife to organize joint planning sequences and negotiate deontic rights (a participants' entitlement to initiate planning sequences and the entitlement to accept or reject a plan). We analyze two different conversations between a man with aphasia and his wife and their adult daughter. Using Conversation Analysis (CA), we identify practices that further the PWA's participation in the interaction while planning afternoon activities together with his wife. The PWA contributes to the planning talk by initiating and modifying planning sequences. The spouse supports his participation by aligning with his initiated actions and inviting him to collaborate in planning talk she initiates. Deontic authority is shared between the conversation partners and the PWA's agency is facilitated even during disagreement. The analysis offers insight into practices that allow a PWA to use his limited communicative resources to contribute competently to planning talk. © 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).


Introduction
Joint planning is a typical activity in everyday life: we plan what to have for dinner, who should pick up the children, when to meet with our friends, etc. It belongs to the family of projective projects in which a speaker attempts to bring about a future action or event -in contrast to reconstructive projects, which deal with the past (Couper-Kuhlen, 2014). In the beginning of joint planning talk, a future scenario might be projected by launching a general idea (Leyland, 2016). Following this, participants collaboratively shape the idea with their actions (Goodwin, 2013). In order to reach an agreement, they modify and negotiate details of the plan step-by-step and make various decisions along the way (Ayaß, 2020). This process is not predefined and can have various formats (Suchman, 2007). Previous research has concentrated on different interrelated elements of joint planning talk: for example proposals (Houtkoop-Steenstra, 1987;Lindstr€ om, 2017), decision making (Huisman, 2001) and negotiation (Siitonen and Wahlberg, 2015). These studies show that one, two or several persons can be involved in planning talk and in taking a decision over a plan (Ayaß, 2020). Furthermore, a plan can be implicative for the person that makes the plan and/or for others (see also Couper-Kuhlen, 2014 about self-and other-agentivity). Additionally, a planning process and the execution of a plan can follow each other immediately or they can be remote by hours, days, weeks or years. If immediate, acceptance of the plan and commitment to it is displayed right away by executing the planned action. If remote, only a commitment to a plan can be claimed because the execution comes about remotely (Houtkoop-Steenstra, 1987). The present study expands the investigation of so-called remote proposals, in which one speaker proposes a future course of action that affects both the speaker and the recipient.
Although the shaping of planning processes is situated in interactions that happen here and now, these interactions influence a person's future lifeworld (Ayaß, 2020). Persons with aphasia 1 are often excluded from planning talk (Johansson et al., 2012) because decision-making capacities are considered to be impaired (Kagan, 1995) and aphasia can mask competence in conversation (Berg et al., 2020;Kagan, 1995). While there is research about participation in planning talk it focuses on PWAs' inclusion in decision making processes in speech and language rehabilitation (Berg et al., 2016;Isaksen, 2018) and hospital settings (Kagan et al., 2020;Simmons-Mackie et al., 2007). Such institutional interactions (in contrast to family settings), have an asymmetrical character, which results in a different pattern of opportunities for participation in decision making processes. Few conversation analytic studies investigate PWAs' participation in planning talk or decision making in familiar settings (Barnes, 2012;Goodwin, 1995). Barnes (2012) shows how a friend supports the planning of a holiday trip by a PWA. Goodwin (1995) describes how a PWA, his wife and a nurse construct what the PWA wants to eat. In both Barnes' (2012) and Goodwin's (1995) studies, the conversation partners drive the organization of planning talk and involve the PWAs by asking questions. The PWAs take a responsive role.
Other conversation analytic research has studied the organization of planning talk for persons without a communication disability in different environments, e.g. for children (Gauvain and Huard, 1999), in the workplace (Greer and Leyland, 2018;Smith, 2005;Stallard, 2000), and for second language learners (Kunitz, 2015;Lee and Burch, 2017;Markee and Kunitz, 2013)). Until now planning talk in PWAs' home environments has not been investigated systematically despite the need to understand which practices facilitate PWAs' participation.
The present study aims to identify practices that further participation by PWAs by analyzing the organization of joint planning sequences in two different face-to-face family interactions, one sequence initiated by a PWA and one by his spouse. It contributes to conversation analytic research on how PWAs and their interlocutors handle and overcome challenges associated with aphasia in interaction and how a PWA's participation may be enhanced (e.g. Barnes and Ferguson, 2012;Bauer, 2009;Beeke et al., 2020;Laakso and Godt, 2016;Lind, 2005). The study adds to our understanding of planning processes and the way in which they allow a PWA to shape his future.

Joint planning talk
An initiation of planning talk can have various forms (Houtkoop-Steenstra, 1987, 1990): a "plan initiator" (Scholtens, 1991:39) can for example make a direct proposal (Let's have macaroni for dinner) (Stevanovic, 2015) or ask an indirect question (What are you doing today?). These forms are united by the fact that they attempt to influence a future activity. Furthermore, regardless their linguistic form, co-participants, the plan recipients, orient to them as a first action by accepting or resisting them (Stevanovic and Per€ akyl€ a, 2012). Thus, the sequential structure of planning talk is founded on a proposalresponse sequence: a first-pair-part of an adjacency pair (Initiation/Proposal) makes necessary a second pair part (Acceptance) (Ekberg, 2011;Houtkoop-Steenstra, 1987). 2 The basic sequential structure of planning talk is characterized as follows for the present study: A: Initiation, B: Acceptance.
Insertion sequences can expand this structure and it can be repeated for negotiating, clarifying, specifying and modifying details of a plan (Ayaß, 2020;Mazeland, 2020). Furthermore, repair can alter or delay the achievement of the pair parts (Clark and Schaeffer, 1989). However, the exact sequential structure of longer planning sequences has received little attention in previous research. The present study takes the described analytical concept of the sequential organization of planning talk as an adjacency pair and focuses on participants' attempts to influence a joint future activity and co-participants' orientation to these attempts.
Planning sequences (initiations and orientations to them) differ in so far as they may support or restrain participation in planning talk. As will be shown, they differ in the degree of (1) deontic authority e participants show who is entitled to initiate planning talk and who is entitled to accept or reject the plan (Asmuß and Oshima, 2012;Stevanovic and Per€ akyl€ a, 2012;Stivers et al., 2018) and (2) co-construction e participants may collaborate or resist in co-constructing the sequence (Stevanovic, 2012). Scrutinizing the participants' practices pertaining to these two concepts may reveal how participation in planning talk can be enabled or challenged.
In line with the first concept, deontic authority, the format of the initiation of a plan (a) shows the degree of entitlement the plan initiator claims to have to initiate a plan, and (b) demonstrates which rights the plan initiator grants the plan recipient to shape a future action 3 (Asmuß and Oshima, 2012;Stevanovic and Per€ akyl€ a, 2012;Stivers et al., 2018). In their study of treatment planning, Stivers et al. (2018) identify four different formats (pronouncement, offer, proposal and suggestion) that may up-or downgrade participants' deontic authority and result in more or less flexibility to jointly plan a future action. 4 With a pronouncement (We are going to), a participant claims high entitlement to initiate a plan and exclusive rights to determine the future. The recipient is not offered the option to decline or negotiate the order, and this restricts the opportunities to collaborate in shaping the plan. By contrast, an offer (Would you like to) expresses low entitlement to initiate a plan and low rights to determine the future. Here, the initiator attributes the decision to the recipient. With a proposal (Why don't we/We can) or a suggestion (You could try/I would), an initiator expresses moderate entitlement to initiate a plan as the plan is presented as optional and not yet determined. While in a proposal, shared rights to determine the future are proposed because the recipient is invited to collaborate in planning the future action, in a suggestion, the initiator claims lower rights to determine the shape of the future action because it invites and depends on the recipients' acceptance. Opportunities for collaboration in planning talk thus appear to be influenced by the design of the initiation.
However, research shows that the distribution of deontic authority can be negotiated as it depends not only on the initiator's claim of rights but also on the recipient's acceptance of the proposed rights (Asmuß and Oshima, 2012;Stevanovic and Per€ akyl€ a, 2012). With regard to proposals in which an initiator proposes shared deontic rights (We can go for a walk later), Stevanovic and Per€ akyl€ a (2012) show that a recipient can confirm the claimed distribution of rights (Yes, let's do that). Yet, she can undermine the proposed distribution of rights by agreeing with the plan but claiming more rights to shape the future action than were granted to her by the initiator (Well, we will certainly go for a walk later). In a study of planning talk involving a PWA, Barnes (2012) finds that the man expresses his rights to a plan by recycling parts of his interlocutors' proposal, question or assertion, and thereby shows his firstness, which is another way of claiming rights to a plan.
Another means to negotiate deontic authority is turn allocation. At each transition-relevant place (TRP) (at the end of each turn unit) a current speaker can continue speaking, select another speaker, or another speaker may select him or herself (Sacks et al., 1974). By continuing to speak or self-selecting, speakers may claim rights to a plan. Shared rights may be offered by selecting a next speaker. Few studies of aphasia have directly considered turn allocation but there are interesting suggestions that aphasia type may have an influence. For example, Ferguson (1998) shows how two persons with fluent aphasia more often select to continue to speak at a TRP than they select their conversation partners without a communication disability as next speakers. In fluent aphasia, 5 sentence structure is relatively intact but neologisms (non-words) and paraphasias (words that sound like or mean something similar to the intended word) are common and logorrhea or 'press of speech' is described, whereby speakers appear to monopolize the conversational floor (Marshall, 2017). In addition, Ferguson (1998) finds these conversation partners do not self-select as next speakers but rather select the PWA as next speaker. However, this finding might be influenced by the institutional relationship between the PWAs and their conversation partners (speech and language therapist, aphasia researcher) or a lack of experience in interacting with PWAs (unfamiliar conversation partners). In contrast, Barnes and Ferguson (2012) find that a familiar conversation partner does not promote the speakership of a person with nonfluent aphasia. In non-fluent aphasia, speech production is halting and effortful and sentence structure is limited although content words may be preserved (Menn et al., 1995). To what extent the type of aphasia influences speakership and thus affects deontic authority remains unknown. While PWAs' limited linguistic resources may create a barrier for verbal organization of speakership, non-verbal resources can be employed to organize speakership and thus facilitate PWAs' authority (Bauer, 2009;Goodwin, 1995;Killmer et al., 2021), as is the case in interactions of persons without communication ability (Mondada, 2016;Rossano, 2012;Zima et al., 2019).
According to the second concept, co-construction of sequences, participation in planning talk becomes visible in participants' collaborative practices while constructing a planning sequence. Stevanovic (2012) identified three subsequent components that establish joint planning talk after an initiation of a plan (e.g. A: "Let's eat ice cream at Ben & Jerry's"). First, the plan recipient shows access or understanding of the plan as B: "They have fantastic ice cream". Then she agrees (B: "I was thinking the same") and finally commits to the plan (B:"Let's go there"). Stevanovic (2012) characterizes decisions in planning talk as collaboratively constructed when the plan recipient develops a plan further through adding an access, agreement and commitment component to a proposal. In this case, participants pursue planning talk by adding attempts to modify or specify the plan, which again can be negotiated collaboratively (Mazeland, 2020). Conversely, if a recipient does not pursue a planning initiation, collaboration in the construction of planning talk may be impeded. Furthermore, planning talk can be constructed collaboratively by expanding an initial idea.
Although previous studies have not considered collaborative construction during planning sequences in aphasia, some reveal more generally how aphasic conversational practices can both promote and impede collaboration. On the one hand, persons with fluent and non-fluent aphasia have difficulties with sequential placement of topic initiations (Barnes et al., 2013) as well as with the lexical composition of sequences that introduce new topics (Barnes et al., 2013;Bauer, 2009;Beeke et al., 2011;Wilkinson, 1999). On the other hand, PWAs' conversation partners can have difficulties with interpreting how a PWA's turn is linked to the previous one (Wilkinson, 1999). Furthermore, Barnes and Ferguson (2012) and Simmons-Mackie and Kagan (1999) find that the sequence initiations of a PWA may not be pursued by the interlocutor. This practice sequentially deletes the PWA's initiations so collaboration in sequence construction is not possible. In contrast, Killmer et al. (2021) show how a conversation partner pursues a PWA's actions by expanding the PWA's talk with and-prefaced turns. In summary, both fluent and non-fluent aphasia may hinder topic initiation, while fluent aphasia can but does not need to hinder a conversation partners' uptake of PWAs' topic initiation e probably depending on the conversation partner.
Another way to co-construct planning talk is to propose new activities during ongoing planning talk conversations. Isaksen (2018) reports that in therapy planning sequences, only speech-language pathologists (SLPs) make proposals and not PWAs. This finding might be influenced by the institutional character of the interaction. However, all SLPs involved in the study stated that they wanted to involve PWAs in decision making and that they encouraged involvement. Isaksen (2018) describes an interaction in which a PWA's initiation of planning sequences is treated as inappropriate by the SLP, who gives minimal 5 The participant in this study has fluent aphasia. responses and does not align with the PWA's initiated action. This asymmetry in initiating proposals during planning talk can also be observed in Barnes' (2012) data. Only the friend launches proposals, not the PWA. Such practices as described by Isaksen (2018) and Barnes (2012) predefine sequences and restrict opportunities for collaboration in constructing planning talk.

The present study
The present study investigates how participation is achieved in the joint planning talk of a PWA and his spouse. More specifically, we examine how the couple constructs two joint planning sequences, with a focus on deontic authority and coconstruction of sequences, which may influence participation in planning talk by supporting or preventing collaboration. We analyze conversational practices that the PWA and his spouse apply to shape actions, and the influence of speaker initiation on planning talk (PWA-initiated planning sequence vs. spouse-initiated planning sequence). It seeks to answer the following questions: (1) How do the participants organize joint planning talk while negotiating deontic rights and what are the consequences for the PWA's participation? (2) How do the sequence initiations (PWA-initiated planning sequence vs. Spouseinitiated planning sequence) influence the organization of the sequence or the accomplishment of participation?

Participants
The participants are a man with aphasia (Fritz, 64 years old), his wife (Helga, 62 years old) and their adult daughter (Uta, age unknown) who is visiting them. 6 Shortly after Fritz retired from his occupation as a chief engineer, he had a stroke. This stroke caused a severe Wernicke's aphasia, a type of fluent aphasia, affecting both his receptive and expressive language modalities (Edwards, 2005; Greenwald, 2018).

Data
The analyzed data originate from the research project Adaptationsstrategien in der famili€ aren Kommunikation zwischen Aphasikern und ihren Partnerinnen (2000e2005) (Adaption strategies in familial communication between aphasics and their partners) directed by Prof. Dr. Peter Auer, University of Freiburg, Germany and Dr. Angelika Bauer, School of Speech and Language Therapy, Freiburg. Germany. 7 The database consists of 142 recordings (in total circa 150 h) and corresponding transcripts of nine German-speaking PWAs who were asked to video record themselves during different typical interactions at home. 8 For the purpose of the present study, the recordings were inspected with a data-driven conversation analytic approach to identify interesting phenomena of the data (Hutchby and Wooffitt, 2008). Joint planning talk emerged as a topic of interest with the potential to give insights into the collaborative organization of participation. Planning sequences by four PWAs were examined in 10 conversations. At this stage, a planning sequence was operationalized as a sequence in which a future activity was proposed by one of the participants. Following this, the analytic focus was narrowed by inspecting sequences in which proposed activities jointly involved the PWA and another participant. Narrowing the focus to joint activities was regarded as creating potential symmetry between the participants because sharing authority in an activity that concerns both participants creates symmetry, while sharing authority in an activity that only concerns one of the participants creates asymmetry. Subsequently, these sequences were categorized according to whether planning talk was immediate or remote. The analysis here focuses on remote planning sequences only, because they involved more talk than immediate planning sequences. Thus, they seemed more challenging for PWAs and provided more opportunities to investigate participation in planning talk. Two remote planning sequences were chosen for analysis here, firstly because they are representative of patterns that the speakers employ across planning sequences. Secondly, they involve different participants as initiators (one sequence is initiated by Fritz and one by his wife). Thereby, they provide a contrast concerning questions of deontic authority. Third, the two sequences reveal successful collaboration, which has the power to provide us with insights into practices that enhance participation. The two sequences are taken from two different recordings. The first recording is 22 min long and involves Fritz, Helga and Uta. We analyzed a 2-min-long sequence that starts 17 min into the recording (Extracts 1, 2 & 3). The second recording is 20 min long and only Fritz and Helga are present. The analyzed sequence starts 7 min into it and is 5 min long (Extracts 4, 5 & 6).
The original transcripts were re-transcribed according to the Gespr€ achsanalytisches Transkriptionssystem (GAT) conventions (Selting et al., 1998(Selting et al., , 2009) (see Appendix for conventions), and translated into English. A multimodal transcription (Mondada, 2006) of the data is added, when of analytical interest. Fritz, Helga and Uta are designated as F, H and U. Because Fritz occasionally produces neologisms (sound strings that are non-words), a gloss line is inserted if relevant, in which unintelligible word forms are transcribed according to German orthography and marked with curly brackets and, if possible, targets of these word forms are provided (Laakso and Godt, 2016). 6 All data presented are anonymized. Names are pseudonyms. 7 For further information about the data see Bauer (2009). 8 Ethical approval for the present study was obtained from the Norwegian Centre for Research Data (NSD).

Analysis
The analysis is divided into two parts, first a planning sequence initiated by Fritz, and then one initiated by Helga. Both sequences can be divided into three phases: a main planning phase and two post expansions (Schegloff, 2007); here called modification phases. All analyzed sequences take place in Fritz and Helga's dining room. The participants are sitting around an oval table with Fritz seated at one end. To Fritz' right is Helga, and Uta (present in the first sequence) is to his left. The camera is placed opposite Fritz.

PWA-initiated planning sequence
In sequence 1 Fritz, Helga and Uta are drinking coffee and eating cake. Fritz has initiated talk about plans for the afternoon 5 min before this sequence by asking Helga "Tja was machsch denn heut du was machen" (well what are you going to do today what to do). Subsequently, the participants talk about different activities they want do in the afternoon when visiting the village of Sankt M€ argen. In this article, we will not analyze how the planning talk was introduced in the conversation as such but instead focus on how the participants launch new plans for specific activities within this ongoing planning talk and thereby initiate new planning sequences. Extract 1a starts when Fritz initiates planning of a new activity, namely to play Lotto (lines 1e4). His initiation is shaped by word misselections (saying "morgen" (tomorrow) instead of "heute nachmittag" (this afternoon)) and a rather vague reference to playing Lotto, namely "eine Serie fahren" (drive a series). This impedes intersubjectivity and leads to a prolonged repair sequence (lines 5e50). We only present the initiation and the end of the repair sequence, as an analysis of the whole sequence would go beyond the focus of the present study. Following this, Helga returns to the planning talk (line 53) and the participants decide to play Lotto (lines 55e59). In the extract we show how Fritz manages to initiate a planning sequence, although topic shift in general is often challenging for PWAs (e.g. Barnes et al., 2013).
Extract 1a e Lotto, Main planning phase The planning sequence is initiated by tying it to the previous plans for the afternoon through and-prefacing (line 1) and the expression "auch noch" (at the same time) (line 2). A future activity is suggested with the time frame "morgen" (tomorrow) (line 1) (Boden, 1997) and the consistency marker "wenn … dann" (when … then) (lines 1 & 2) (Mazeland, 2020). "Morgen" (tomorrow), whilst appearing to be a word misselection (the trip is established as occurring this afternoon) still serves to indicate a future point in time. The sequence is thus launched in a turn-design that serves as a transition from previous planning activities to the introduction of a new activity.
Although it is rather unclear exactly what Fritz is proposing to do (more on this below), his utterance is recognizable as a proposal for some future action involving Helga. By launching a planning sequence in a self-selection, Fritz claims rights to initiate a plan, and thereby claims some degree of deontic authority. The modal verb "kann" (can) (line 2) expresses modal possibility (Klabunde, 2007), which claims moderate entitlement and shared rights to determine the future action. Furthermore, Fritz downgrades the contingencies of the plan by the use of the minimizer "a bissl" (a bit) (line 2), presenting the plan as less intrusive and orienting to Helga's perspective on the plan. In this way, the format of the initiation displays the plan as not yet settled and open for further discussion.
Before analyzing the uptake of the proposal, we need to understand how the interlocutors achieve intersubjectivity on what is being proposed. In line 4, it is not clear whether Fritz says "See rüber fahren" (to drive across the lake) or "Serie fahren" (to drive a series). If the former, it could be a suggestion to drive across a lake on the way to Sankt M€ argen. The couple lives in an area with some lakes. If it is the latter, Fritz could be proposing to fill in a couple of Lotto tickets and play a series of Lotto games. In Germany, it is common to fill in a couple of tickets at once to play for a whole month in advance. Yet, it is not conventionally referred to as "Serie fahren" (to drive a series). Thus, this could also indicate a word misselection.
In making the proposal, Fritz slides his left index finger over the table to the left and to the right (line 4). This occurs after a pause and starts while he is talking about the activity. It has been observed previously that PWAs use more gestures than their interlocutors (Auer and Bauer, 2011), and this gesture may help us understand what he is trying to say. The same gesture occurs in line 46 while uttering "Lotto", so it is likely that it illustrates the same activity. This would support the suggestion that he says "Serie fahren" (to drive a series). Thus, the proposed activity is embodied in a pre-enactment (Leyland, 2016) and a future activity is made visible (Lilja and Piirainen-Marsh, 2019). However, at this point Helga initiates repair by opening her eyes wider and leaning forward (Mortensen, 2012) (line 5) and a prolonged repair sequence follows. Intersubjectivity is restored when Helga displays her understanding in a series of understanding checks (lines 44e53) (Jefferson, 1972). Subsequently, she makes a self-initiated return to the planning activity by producing a response to the original proposal (line 55).
In her response, Helga confirms the proposed distribution of deontic rights, shows access to the proposal (line 44), agrees to it and commits to the plan (line 55). The affirmative response token "ja" (yes) displays agreement and compliance to the suggestion. She also expresses her commitment to the plan by repeating the suggestion: "k€ onnt mer eigentlich" (we could actually). By using modal force of possibility, she hands the decision back to Fritz. In addition, she uses the pronoun "merwir" (we), thereby displaying her understanding of Fritz' utterance as a proposal for a joint activity. Thus, also in constructing Fritz as a co-agent of the future action (Bauer, 2009;Couper-Kuhlen, 2014), she expresses their shared rights to decide on the plan. She is orienting to him as partner and husband with equal rights, which might reflect the dynamics of this couple in terms of decision making that go beyond issues related to the aphasia.
The main phase of this planning sequence thus contains a proposal (lines 1e4) and three subsequent components of planning talk as discussed by Stevanovic (2012); access (line 44), agreement and commitment to the plan (both line 55), and these build on each other. With regard to co-construction, we see that Fritz and Helga collaboratively accomplish the planning activity. Fritz initiates the sequence with a proposal that invites Helga's collaboration (lines 1e4). With this initiation, he creates the opportunity to participate in planning talk for himself while giving Helga a chance to co-shape the plan. After the repair sequence, Helga returns to the activity of planning, thereby constructing Fritz' initiative as important and consequential. This is counter to what has been observed in previous studies on uptake of PWAs' initiatives in interactions with familiar interlocutors (e.g. Barnes and Ferguson, 2012) and shows joint accountability for the continuation of the action of planning.
In the next phase, presented in the Extract 1b below, Helga suggests a modification of the plan, namely that Fritz should play Lotto at Martha's (likely the owner of the place where they want to play Lotto) (line 46). All participants agree to this suggestion (lines 61e67). Extract 1b, Lotto e Modification 1 Helga proposes a modification of the plan by suggesting that it is Fritz (and not her) who should play Lotto and that they can do it at Martha's. She initiates the modification of the plan as a pronouncement, telling Fritz what he should do: "kannsch DU des mache" (YOU can do it, line 60). This claim of high entitlement and unilateral authority needs to be seen in context. Since the participants have already agreed on the overarching plan to play Lotto, they have established joint commitment to a common goal and need not orient to contingencies to the same degree (Rossi, 2012). The claim of deontic authority is also counterbalanced by other aspects of the turn design, which orient to Fritz' rights to contribute to shaping the plan and deciding. First, the pronouncement starts with the consistency marker "dann" (then) (Mazeland, 2020), referring back to Fritz' initiation and showing that it is based on his initiative. Furthermore, the modal verb "kannsch" (you can) presents this proposal as a merely a possibility and thus contingent on Fritz' acceptance. Finally, she downgrades the deontic force by adding a tag question "ne?" (right), inviting Fritz to collaborate (line 62).
Requests for remote action prefer expanded responses with an explicit statement of commitment (Houtkoop-Steenstra, 1987;Lindstr€ om, 1999Lindstr€ om, , 2017. Fritz' first response (line 61) is not treated as sufficient, and a more substantial response is pursued by the tag question. At this point, Fritz gives the preferred response by agreeing "ha ja" (well yes) (line 63), evaluating the plan a positive "gut" (good) (line 63), showing access to it "von da aus" (from that perspective) (line 987) and committing to the plan "mach ich des" (I will do that) (line 65). Similarly to Extract 1a, the phase is closed after all participants have agreed to the modification (lines 65e67). Again, Fritz and Helga engage in this phase, and Uta aligns.
Fritz introduces a second modification in Extract 1c (lines 69e75) by asking his daughter to help him play Lotto. With this introduction, he provides new opportunities for himself to shape the plan and thus to participate in planning talk.

Extract 1c, Lotto e Modification 2
Fritz claims high entitlement to modify the plan and unilateral rights to determine the future action by requesting help from his daughter in a pronouncement (line 72). He addresses Uta by pointing and looking at her (line 69) and by using the pronoun "du" (you) (line 72). The modal verb "musch" (have to) (line 72) expresses necessity and thus claims high entitlement to initiate the modification and unilateral rights to determine the future action. Once again, the established joint commitment to the common goal legitimizes not orienting much to contingencies (Rossi 2012). As is typical for pronouncements such as this one, the request is accompanied by accounts for his need for help, namely that he does not know how to play it e probably because the aphasia hinders him (lines 69, 74 & 75) (Ekberg, 2011;Houtkoop-Steenstra, 1987).
Fritz' employs gaze as a means to modify deontic rights between participants. He shifts his gaze between Helga and Uta during three different attempts to mobilize help when playing Lotto. During Fritz' first attempt (lines 69e71), he addresses Uta by gazing and pointing at her. Uta does not respond: she sits still and leans on her left hand with the elbow on the table and fingers in front of her mouth. During Fritz' second attempt (line 72), he shifts his gaze to Helga at the end of his turn to requests a response from her. After getting no response (pause in line 73), Fritz makes a third attempt by reinforcing the need for help playing Lotto while shifting his gaze direction back to Uta (lines 74 & 75).
Prompted by Fritz' gaze behavior, Helga responds on their daughter's behalf and pronounces that he will do it with Uta's directions (line 77), thereby taking her compliance for granted. Uta responds to this with just a weak agreement (line 79), without providing an independent expression of access and commitment, as would be the expected response. In this way, Helga, who up until this point has been sharing rights to make plans with her husband Fritz, denies the same rights to her daughter, Uta. By speaking on her behalf and presupposing compliance, she treats her as not having a part in the activity of planning, perhaps orienting to her as a daughter (although she is grown-up), expected to comply with her parents' demands.
The Lotto planning sequence closes after this second modification. They go on to joke about Fritz and Uta having to share the potential Lotto prize in case Uta assists in filling in the coupon.
In this planning sequence initiated by Fritz, we have shown that the participants accomplish joint participation by coconstructing the planning sequence, up-and downgrading rights and inviting collaboration. In order understand the importance of the role of being the initiator of a planning activity we will compare this extract to one where Helga initiates a planning sequence.

Interlocutor-initiated planning sequence
Prior to the next extract, Fritz and Helga talk about the current weather conditions. It is a sunny day and there is snow outside. Helga initiates the planning sequence by proposing to go for a walk in the afternoon (lines 1e3 & 5). Fritz agrees to the plan (lines 4, 6 & 8e9). Then a repair sequence (lines 10e36) and talk about related topics follows, including the fact that Fritz shoveled snow in the morning, and the snow conditions on the streets (lines 37e196). Again, we present parts of the repair sequence and related talk only when they are relevant to the analysis of planning talk. In lines 197e201, Fritz initiates a return to the previous planning talk. Then both participants repeat Helga's initial proposal (lines 202e205). The plan is presented as a suggestion, with moderate entitlement to initiate the plan and shared rights to determine the future action. To go for a walk is treated as optional by the modal verb "k€ onne" (can) (line 2) which expresses a possibility (Klabunde, 2007) and invites Fritz' collaboration and acceptance. The minimizer "nommal" (again) (line 2) presents the plan as less intrusive and as a low threshold for Fritz which shows that Helga orients to Fritz' perspective on the plan. Furthermore, with the tag-question "oder" (right) (line 3), Helga accommodates proposals for doing something else and open to Fritz' assessment (Pomerantz, 1984), which increases his deontic rights. Upgrading Fritz' rights at this point could be a reflection of the dynamics of decision making for this couple and not solely about aphasia. A short pause occurs after the tag question (line 3) as Fritz does not give an immediate response. Following, Helga downgrades her own entitlement and upgrades Fritz' rights to determine the plan with "was meinsch du" (what do you think) (line 5). This addition presents the plan as a proposal, which leaves Fritz more discretion in the decision.
Fritz agrees and commits to the plan (lines 4, 6 & 8). The modal verb "k€ onne" (can) (line 4) shows commitment to the plan and confirms Helga's proposed shared deontic rights. Yet, the agreement with the plan is expressed with a particle "ha" (hah) (line 4), which expresses Fritz' high entitlement to the plan. This particle -in contrast to a simple "yes" shows that it is indeed a possibility that they can go for a walk and expresses his rights to the plan (Klabunde, 2007). In addition, he upgrades his deontic authority by providing an independent reason for taking a walk (watching the snow), thus contributing to shaping the proposal rather than merely going along with Helga's suggestion. The sequence ends with three repetitions by Fritz and Helga of the initial proposal to go for a walk, which manifests their mutual agreement to it (lines 202, 203 & 205).
There is a high degree of collaboration in the construction of this planning sequence. Helga explicitly asks Fritz for his opinion and thereby invites him to collaborate in shaping the plan (lines 3 & 5). Fritz accepts the invitation and pursues the construction of the sequence by agreeing to the plan "ha ja" (hah yes) and committing to it: "mr k€ onne de scheg {snow} amal ansehn" (we can once look at the {snow}) (line 4). Due to overlapping talk (lines 4 & 5), Fritz repeats this agreement and the commitment in lines 6 and 8. Helga's "hm?" (line 7) appears to reflect Fritz' incomplete commitment to her remote proposal (Houtkoop-Steenstra, 1987;Lindstr€ om, 1999Lindstr€ om, , 2017 and probably also his missing display of access to the plan. It is not clear whether Fritz has access to the plan due to the neologism "de scheg" (line 4), probably referring to snow. At this point, the sequence could have been closed as the participants have reached mutual agreement. However, Helga initiates repair targeting the neologism "de scheg". After the participants have reached mutual understanding (line 36), they abandon the planning talk in a long stretch of non-planning talk. At a certain point, Fritz re-introduces the previous planning talk in a selfselection (lines 195e201 & 203). He repeats Helga's proposal (line 203) in overlap with Helga who also repeats her proposal (line 202). Fritz substitutes "going for a walk" with "work" here, a substitution that he uses repeatedly in this conversation. Fritz' voluntarily returning to the action of planning talk that Helga initiated strengthens their collaboration in framing the action and shows their joint accountability for the action. Furthermore, by asking Helga whether they want to go for a walk, Fritz passes the right to decide back to Helga, which might reflect the dynamics of decision making for this couple and go beyond issues related to the aphasia.
In Extract 2b, the action of planning continues when Fritz raises the question of what type of shoes they will wear during the walk (line 206e210). He contextualizes his question by pointing to his foot (line 207). After a repair sequence dealing with Fritz' proposal to put on different shoes (lines 211e223), Helga and Fritz negotiate which shoes they should wear and come to the agreement not to wear their snow boots (lines 220e236). Just as in sequence 1, this planning sequence is constructed collaboratively as Fritz continues the action with a specification of the plan proposed by Helga. Yet, in this case, they have to deal with a disagreement and negotiate a solution.
Extract 2b, Walk e Modification 1 Fritz launches the modification with a question that seeks to specify the plan (lines 206e208), thereby self-selecting to contribute to shaping the plan. Fritz downgrades his own rights and upgrades Helga's by allocating the decision to Helga (line 206), which might be a reflection of the dynamics of this couple in terms of decision making and not solely about the aphasia. The question is first formulated as an open question but instantly reformulated by proposing two candidate alternatives. The question is biased towards one of the alternatives in that he mentions a problem with using the shoes they wore this afternoon, namely that they are wet. After a repair sequence and a series of understanding checks by Helga (lines 220 & 221, and 229), they establish that the other alternative he is suggesting is to wear snow boots. Sometimes, repair initiations may be taken to adumbrate disagreement (Schegloff, 2007;Svennevig, 2008), and so may requests for clarification such as the one in line 225 (Pomerantz, 1984). Fritz seems to orient to this contingency in that he reformulates the question with reversed preference structure (line 231) (Sacks, 1987). By doing so, he once again explicitly allocates the decision to Helga. Helga, on her side, declines to take the decision for them both. Whereas Fritz included Helga in his initial question by referring to both of them with the first person plural pronoun "mr" (we) (line 206), Helga's answer only refers to herself with the first person singular pronoun "mr" (for me). 9 In addition, she mitigates the deontic force of her statement with the epistemic hedge "ich glaub" (I think) (line 233). Consequently, she reduces her deontic authority over the plan by stressing that she only speaks on her own behalf.
Fritz only acknowledges Helga's final account with a relatively noncommittal acknowledgement token uttered in a soft voice in the end of this phase (line 234) (Jefferson, 1983) instead of a stronger agreement and commitment. This signals less than full acceptance of the solution, and, as we shall see, foreshadows disagreement, which he makes more explicit in the next extract (lines 237e242). Consequently, the negotiation continues (lines 253e253). Eventually, the phase and thereby the whole sequence end with agreement not to wear snow boats now but rather when it gets colder (lines 254e261).
Extract 2c, Walk e Modification 2 Both partners here back their position with accounts. First, Fritz returns to his proposal to put on snow boots by giving an account for why other shoes are insufficient. (lines 237e242). Framing the account with "ich denke" (I think), mitigates the disagreement. Subsequently, Helga gives an account for why the other shoes are sufficient (lines 243e246) and Fritz expresses agreement (lines 247 & 248). Fritz' response "ja" (yes) (line 247) while looking at Helga and then at the table may merely be an acknowledgement and "((clears throat)) ja na GUT" (yes well good) (line 248) while looking outside the window may be an attempt to close the sequence. However, the fact that Fritz does not continue to negotiate shows agreement to some point here. Both partners express high entitlement to negotiate the plan by pursuing compliance to their original position. On the other hand, giving accounts displays that they take the other party's position seriously and seek to convince them by arguments. Orienting to each other as equal partners, which one takes serious could reflect the dynamics of this couple in terms of decision making that go beyond issues related to the aphasia.
After Fritz has expressed acceptance, Helga continues to give further accounts for not choosing snow boots (lines 249e256). This pursuit suggests that she seeks a more explicit statement of commitment (Houtkoop-Steenstra, 1987;Lindstr€ om, 1999Lindstr€ om, , 2017. This pursuit may be considered successful in that Fritz does commit to the plan by repeating (in a slightly modified form) Helga's decision to wait until it gets colder (line 257). In lines 258e260, Helga settles the agreement with an affirmation token and yet a modified repeat. After a pause (line 261), Fritz initiates a new topic and the planning sequence is closed.
In the two modification phases studied here, we see that also disagreement between the parties is handled in a way that distributes the deontic rights between the parties rather equally and allows negotiation of a shared decision. Both parties balance between asserting their point of view and providing room for the other to express their opposing views. Even though it is the spouse who 'wins the argument', the agreement is based on the strength of their respective arguments rather than on their communicative abilities to assert them.

Discussion and conclusion
The present study illustrates how a man with severe fluent aphasia is able to participate in planning talk. It explores practices deployed by the PWA and his spouse to coordinate joint planning sequences. Thereby, the study contributes to our understanding of planning processes and the way in which these allow a speaker with aphasia to shape his future.
The first research question concerned the organization of the planning sequences. These data show two important practices for organizing planning sequences and facilitating the PWA's participation in them. The first one is PWA's initiation of a planning sequence within ongoing planning talk. The second one is participants' collaboration in constructing a planning sequence, which is enabled by the participants inviting each other to collaborate. The PWA is constructed as a competent interlocutor and a legitimate decision-maker by his wife who actively invites collaboration as well as leaving space on the conversational floor. This is in contrast to previous studies that describe the exclusion of PWAs during planning talk (Johansson et al., 2012).
During planning talk (for example planning what to do this afternoon), the participants propose different activities/topics in planning sequences. These sequences are divided into a main planning phase of the overall activity (for example playing Lotto) and a modification phase where details are coordinated (for example where to play Lotto). All phases are organized in adjacency pair-sequences. Overall, the structural organization is similar to planning talk in typical interactions (e.g. Ayaß, 2020;Ekberg, 2011;Houtkoop-Steenstra, 1987;Mazeland, 2020) apart from the presence of frequent repair sequences, which reflect the atypical character of the interaction (Bauer, 2009).
The second research question sought to compare the influence of speaker initiation on the practices of planning sequences (PWA-initiated vs. spouse-initiated). The practices are similar in these data. In both analyzed sequences, regardless of who initiates the planning talk, both participants negotiate plans and both have the right to disagree. In this way, planning talk is collaboratively constructed. This enables the PWA to influence a future activity. In these data, the organization of planning talk is not predefined. All participants equally shape it, which is in contrast to planning sequences described in previous studies, in which PWA's interlocutor unilaterally moderates the interaction (Barnes, 2012;Goodwin, 1995;Isaksen, 2018 ,10 ).
Deontic rights are essential when negotiating authority over a plan. The PWA claims deontic rights in the first place by launching a planning sequence within ongoing planning talk and initiating modifications in an other-initiated planning sequence. This practice displays his rights to initiate and determine future activities. Such self-selected initiations have seldom been documented in planning talk of PWA (Barnes, 2012;Isaksen, 2018 ,10 ).
The participants employ different practices to upgrade or downgrade deontic rights. The PWA claims rights in a pronouncement with the construction "you have to" or allocates rights to his conversation partner by asking an open question, giving alternatives, or asking what the conversation partner wants. Additionally, the PWA uses non-verbal resources such as gaze to modify rights between participants, a practice that has been described in previous studies (Killmer et al., 2021). The conversation partner downgrades her own rights by adding the tag question "right" after her proposals, and by asking explicitly for the PWA's opinion "what do you think". Accordingly, the participants regulate authority in planning talk by upgrading and downgrading their own and their interlocutor's entitlement to the plan.
Furthermore, when initiating plans both partners propose shared distribution of rights to determine the future action. And as recipients, they confirm the proposed distribution of rights. By presenting plans as proposals with the modal verbs "can" or "could" both interlocutors propose shared rights by presenting plans as not settled and open for further discussion. In addition, the PWA's conversation partner proposes to share rights by constructing the PWA as co-agent of the future action (with the personal pronoun "we" (Bauer, 2009;Couper-Kuhlen, 2014). These practices have the effect of promoting the sharing of rights throughout the planning talk. However, after the interlocutors have agreed on a main plan, modifications may be initiated by claiming higher rights to them. Thus, both partners make strong claims subsequently to the establishment of a joint project with a common goal (Rossi, 2012).
Co-construction is a central approach when creating planning sequences in our data. Both conversation partners progress each other's actions. They do this by, on the one hand, pursuing an action through initiating a next component of a phase (access, agreement and commitment). On the other hand, they add modifications to main planning talk initiated by the other person. By pursuing the action, each of them displays that they are working on the same project. PWA's turns appear wellfitted to turn taking in terms of initiations and modifications of planning talk. Thus, the speakership changes are in accordance with the requirements of the planning activity. Furthermore, the conversation partner does not sequentially delete PWA's turns. This is in line with descriptions of conversational strategies to pursue a PWA's project (for example Killmer et al., 2021). However it differs from studies showing that interlocutors did not pursue a PWA's actions (Barnes and Ferguson, 2012;Simmons-Mackie and Kagan, 1999).
Different practices are employed to enable co-construction and further collaboration. The PWA's initiation of actions is nicely embedded in the sequential context and makes collaboration possible. He launches a new planning sequence using a turn-design that connects it to the ongoing planning talk with and-prefacing, and by indicating a future time frame (tomorrow). Furthermore, he promotes his proposals with non-verbal resources such as gestures that pre-enact (Leyland, 2016) the future activity (Lilja and Piirainen-Marsh, 2019) or by pointing and looking at his conversation partner. Additionally, by launching plans as proposals and not as pronouncements, he invites collaboration. These results confirm claims that people with fluent and non-fluent aphasia are able to initiate new conversational sequences successfully using distinct turn design practices accompanied by topic shift markers, and deployed with competent sequential timing (Barnes et al., 2013;Bauer, 2009;Beeke et al., 2011;Killmer et al., 2021;Wilkinson, 1999). The conversation partner facilitates collaboration and participation by explicitly asking for the PWA's opinion (what do you think?) and returning to planning talk that was initiated by the PWA after a long repair sequence. These practices emphasize joint accountability for the planning process.
In case of disagreement, both partners express high entitlement to negotiate the plan by pursuing compliance to their original position. During this negotiation process, they give accounts for their point of view, which displays that they take the other party's position seriously and seek to convince them by presenting arguments. Both apply the epistemic hedge "I think" to mitigate the deontic force of a statement. They reduce their deontic authority by stressing that they speak on their own behalf.
Whilst it should be acknowledged that this is an analysis of a conversation involving one individual with aphasia, many of the practices we observed are characteristic of typical planning talk, and therefore it may be the case that other PWAs are able to participate as planning partners in such sequences. In order to strengthen these results, a broader study of planning talk with a range of speakers with differing aphasia types is desirable. Further research into planning talk by people with aphasia will also broaden our perspectives and concepts for speech and language interventions that aim to improve conversations involving a PWA (Beeke et al., 2015;Wilkinson, 2010). Our findings show that communication partner training (CPT) could benefit people with aphasia and their families by facilitating activities such as joint planning talk. As part of CPT, PWAs and their family members could be educated by speech and language therapists (SLTs) about how planning talk works and could reflect on their own joint planning, as a way of enhancing participation in planning talk for the PWA.
In summary, these results display how collaboration secures the active participation of a person with severe fluent aphasia in planning talk, and how genuine joint planning between equal partners is accomplished. We show that collaboration can support planning talk and that interlocutors collaborate despite language difficulties. A PWA can be a competent interactant when she actively initiates planning talk and the conversation partner acknowledges PWA's rights to the plan. The present analysis indicates that interlocutors play an important role in supporting PWA's conversational agency by employing conversational practices that enable PWAs' inclusion in planning processes and make participation in a typical everyday activity possible.

Declaration of competing interest
The authors have no conflicts of interest. interpretive comment with indication of scope ((coughs)) nonverbal vocal actions and events <<coughing> > … with indication of scope ( ) unintelligible passage (xxx), (xxxxxx) one or two unintelligible syllables (may i) assumed wording (may i say/let us say) possible alternatives ((unintelligible, appr. 3 sec)) unintelligible passage with indication of duration ((...)) omission in transcript --> refers to a line of transcript relevant in the argument Additions by present author(s) f:/h:^/* representing non-verbal behavior (e.g. gestures, movements and gaze) ?: unknown speaker