Abstract
Although institutional discourse is subject to a vast ensemble of constraints, its design is not fixed beforehand. On the contrary, optimizing the satisfaction of these constraints requires considerable discourse design skills from institutional agents. In this article, we analyze how Dutch banks’ mortgage advisors navigate their way through the consultations context. We focus on what we call discourse design explications, that is, stretches of talk in which participants refer to conflicting constraints in the discourse context, at the same time proposing particular discourse designs for dealing with these conflicts. We start by discussing three forms of design explication. Then we will examine the various resolutions they propose for constraint conflicts and show how advisors seek customer consent or cooperation for the proposed designs. Thus our analysis reveals how institutional agents, while providing services, work on demonstrating how the design of these services is optimized and tailored to customers.
Keywords
Introduction
We are all discourse designers: talking to each other, we design utterances on a routine and on-the-fly basis, taking into account constraints deriving from the talk’s purpose and context. In institutional contexts, managing constraints is even more essential because most of these contexts imply a vast number of discourse constraints. Discourse constraint management is sometimes done explicitly. Excerpt 1 is taken from a Dutch mortgage consultation:
The ultimate purpose of this consultation is to make the customer understand basic mortgage information. However, there arises a dilemma in how to achieve this purpose. In line 4, the advisor (AD) explains that he wants to deliver a ‘sensible story’, which we take to mean that he delivers his usual comprehensive mortgage introduction in order to prevent questions later on and thus to save time; this is an efficiency constraint. Heeding this constraint may, however, lead to the violation of a politeness constraint, as telling things the customers may already know (line 3) may come across as underestimating their knowledge, and hence as being impolite.
In order to resolve the conflict the advisor produces a design explication, that is, an utterance referring to a constraint conflict. At the same time, he proposes his solution: prioritizing the coherence of his story over adjusting it to hearer knowledge. Implicitly, he solicits hearer consent for this course of action. So, this discourse design explication showcases how the advisor maneuvers through a discourse context presenting conflicting constraints.
Such discourse design explications occur regularly in mortgage consultations since mortgage advisors are bound by factors such as internal and external institutional policies, the different interests of various departments (i.e. legal vs marketing) and the different interests of agents (i.e. mortgage advisors and customers).
More generally, design explications display a distinctive feature of discourse design: any design is responsive to contextual pressures, but discourse is special in that talk may represent context and be explicit about the way it responds to contextual pressures. Hence we are dealing with public displays of discourse design-in-the-making.
In this article, we will focus on these explications in the particular context of Dutch mortgage consultations, which we will analyze from three angles. First, they provide us with a window on some of the constraints that are relevant in our particular genre of interaction, and more generally on how the organizational context may affect discourse. Second, they demonstrate how experienced professionals, using design strategies that have been honed over time, deal with discourse options and dilemmas and try to satisfy as many constraints as possible. And third, the explications are interactional moves, showing the advisor’s expertise as well as inviting customers to participate in or at least consent to discourse designs. Let us first discuss our core concepts of constraints and discourse design before moving on to our dataset and the actual analysis.
Discourse constraints
We define a constraint as a limitation to the options available to an interaction participant. Consider the Venn diagram in Figure 1, in which utterance options are represented by asterisks. The largest ellipse shows the set of possible utterances (SPU) theoretically available to a mortgage advisor to achieve a basic consultation purpose, for example, explaining basic mortgage concepts in Excerpt 1. This basic purpose constitutes the first constraint impacting the interaction. This SPU is derived solely from purpose-related constraints. However, various further constraints need to be heeded in the interaction; in Excerpt 1, these were politeness (constraint A in Figure 1) and efficiency (constraint B). These further constraints carve out subsets from the primary SPU: the advisor should manage customers’ preexisting knowledge, otherwise they may get the feeling that the advisor thinks they are stupid. Furthermore, the advisor needs to make sure his story is delivered as efficiently as possible (constraint B), given that time is money.
Constraints narrowing down the purpose set of possible utterances.
We will call these further constraints non-purpose constraints. They derive from different ‘aspect systems’ (Veeke et al., 2008), a notion to be explained later. Supposing that Figure 1 represents the situation of Excerpt 1, there is no utterance that would satisfy all constraints, as the intersection of ellipses A and B contains no asterisk. Hence the advisor can only prioritize one constraint over the other, which is the option actually chosen Excerpt 1.
Given that their optimal next move is often not self-evident, advisors regularly need to engage in discourse design. In other words, constraints by themselves do not produce interactions. They merely provide ‘structure’, in the sense of the structurational analysis of social systems pioneered by Giddens (1984; see Carter and Sealey, 2000, for further discussion): they provide rules and resources that are drawn upon and acted upon by individuals. Social action and social reality only emerge in the interaction between agency and structure, two entities that can never be reduced to one another. More specifically, our analysis of design explications demonstrates how the discourse context not only constrains the advisors’ set of interactional options, but requires them to use professional skills and creativity in navigating their design space. This involves what Giddens (1984) has called reflexive monitoring: ‘In circumstances of interaction – encounters and episodes – the reflexive monitoring of action typically, and routinely, incorporates the monitoring of the setting of such interaction’ (Giddens, 1984: 3). While much of this monitoring goes on implicitly, this article analyzes interactional displays of reflexive monitoring, in which the practical consciousness of institutional actors takes on discursive forms.
We have already distinguished purpose and non-purpose constraints. The non-purpose constraints in Excerpt 1 concern politeness and efficiency, and they derive from different aspect systems impacting the consultation. This notion of aspect systems stems from the Delft systems approach to organizational analysis (Veeke et al., 2008), which postulates that all systems consist of elements (subsystems) linked to each other by different relations (aspect systems). For instance, our mortgage provider, bank B, has a department (i.e. a subsystem) called ‘Mortgage Communication’ that develops several communication products, including the mortgage consultation. These products are further subsystems within the communication department (see the vertical pillars in Figure 2) and may even be visible as such in the organizational structure in the sense that specific working units correspond with the different products.
Potential subsystems and aspect systems relevant to mortgage communication.
Every working unit needs to consider various aspect systems, for example technological, efficiency and politeness aspect systems. These aspects refer to different kinds of conditional relations between activities, providing possibilities and impossibilities. Aspect system issues may concern the technology required to build a website, the time available to talk with customers or the preferred way of approaching customers. Such aspect system issues generally apply to various subsystems simultaneously, so they can be represented as bars ‘crossing’ the subsystem columns (Figure 2).
Our distinction between purpose-based and aspect system-based constraints is a well-known one in design thinking. For instance, software engineers (Chung and Do Prado Leite, 2009; Glinz, 2007) talk about ‘functional’ and ‘non-functional’ requirements of applications. Examples of ‘non-functional’ requirements in software design are speed, physical requirements, security and interface usability. The label is slightly misleading, however, as these constraints are every bit as important as the functional constraints. The same goes for the aspect system constraints in our study, which stem from the entire organizational context. They represent essential conditions for the interaction to take place at all.
Talk as a design practice
The perspective of talk as a design practice has been adopted in various traditions: discourse analysts have used it in work on features such as coding the information status of referents, perspective taking in lexical choice, syntactic organization and prosody (see Fox, 2008); conversation analysts have discussed turn design, action formation and preference organization (see Drew, 2013; Levinson, 2013; Pomerantz and Heritage, 2013, for overviews); within linguistics, Optimality Theory focuses on how utterances satisfy constraints (Prince and Smolensky, 2004) and how hearers make inferences based on the assumption of constraint satisfaction (Hendriks and De Hoop, 2001). According to these traditions, much of the talking design work takes place unconsciously and is primarily focused on adjusting utterances to their recipients and the immediate context. However, O’Keefe (1988) proposed an elaborated analysis of communicators’ assumptions underlying talking design work in order to explain design variations. Her message design logics theory explains differences between language users, which primarily appear when the message context presents complexities, for instance because of bad news.
Aakhus and Jackson (2005) share this interest in researching communicators’ assumptions, but include contextual factors such as technology. For them,
taking a design stance toward such technology includes, at a minimum, seeing what hypothesis about communication is expressed in the design and being able to make reasonable assessments of whether people’s use of the technology is adapted to its design features or struggles against its design flaws. (p. 414)
They state that message designs are constrained by contextual factors that bring their own designs with them; these factors may be technological in nature, but organizational as well.
Face-to-face interaction design is peculiar in that constraints may be articulated and implemented at the same time. This kind of designing-on-the-fly is a less studied phenomenon. One study by Aakhus and Rumsey (2010) reports an interactional discourse design analysis of a disagreement about interactional norms in an online cancer support group forum. The taken-for-granted interactional norms posed an interactional design dilemma because they were challenged by some participants but at the same time used by others to get a derailed complaint situation back on track.
Aakhus and Rumsey (2010) deal with design dilemmas that are largely implicit and need to be reconstructed from the interactional moves of the various participants, while our design dilemmas are referred to more explicitly in the interaction, as our mortgage advisors regularly talk about the various constraints applying to mortgage orientation consultations. Another difference between this study and theirs is that we deal with a more rigidly constrained type of institutional discourse.
In what follows, we will first describe our data, that is, the collection of mortgage consultations and organizational background documents clarifying the consultation context. Subsequently, we will analyze the consultation context in terms of purpose constraints and aspect system constraints. This context analysis helps us to collect discourse design explications in our data. The analysis of the explications starts with discussing three explication forms. Next, we will analyze the constraint conflict solutions proposed in explications. And finally, we will shed light on the interactional role of the explications in this particular type of institutional discourse, especially on how they solicit customer consent and cooperation and how they serve to profile expertise and individual agency given institutional constraints.
Data
Bank B, one of the main mortgage providers in the Netherlands, allowed us to record 39 mortgage orientation consultations, with a length varying from 45 minutes to up to 2 hours. All recordings were orthographically transcribed, thus enabling word searches.
Orientation consultations are just one type of consultations in the Dutch mortgage purchase process; other types are advice consultations and mortgage quote signing consultations. They appear in this order in the mortgage purchase process. Due to Dutch legislation launched in January 2013 (Besluit Gedragstoezicht financiële ondernemingen (BGfo) Wet op het financieel toezicht (WFT), 2013: Article 86C), the orientation consultation is free of charge, but the other ones are not.
In the orientation consultation, advisors explain bank B’s basic mortgage options (mortgage forms, interest rates and interest rate periods), run a (maximum) mortgage loan amount calculation and discuss the outcomes of the calculation. So, the main advisor activities in the orientation consultation are data gathering, data entering and explaining. When customers request advice, advisors cannot fully comply with their request as a result of the Dutch legislation mentioned above.
Mortgage consultations are part of a multichannel communication package (MCP) that supports home buyers (Herijgers and Pander Maat, 2015). We asked the MCP stakeholders to provide us with relevant organizational documents on bank B’s organization structure, on bank B’s communicative and organizational purposes and on other constraints impacting the consultations. These stakeholders provided us with documents, some confidential, and research reports on communication topics such as target groups and mortgage customer journeys. Along with the consultation data, these documents enabled us to reconstruct the consultation context.
Consultation context analysis
We used the consultation corpus and the afore here mentioned organizational documents to identify the purpose constraints and the aspect system constraints. We opted to combine data sources because it is entirely possible that a constraint is operative but does not surface in the explicit form of a discourse design explication. In fact, it is an empirical question as to which constraints tend to be invoked in the interaction and which constraints tend to remain tacit knowledge.
We started by listing potential purpose constraints. In the consultation transcripts, customers present four different reasons to visit bank B’s mortgage advisor (see Table 1, left column). From these, we derived the consultation’s communicative purposes, described in terms of the intended cognitive effect on customers (see Lentz and Pander Maat, 2004). These candidate purposes were validated by checking them with one of bank B’s advisors.
Customer purposes and advisor purposes in the orientation consultation.
First, customers seek generic information on mortgage options and a mortgage purchase in general (Herijgers and Pander Maat, 2015). Second, they want to buy a house and want to know how much money they can borrow. Third, they may have set their sights on a house and want to know whether this particular house is affordable. Fourth, they may have made an actual bid on a house under financing conditions and want to check bank B’s purchase support. This applies to returning customers who left to find a house then come back to check whether their initial maximum mortgage loan amount is still applicable. Interest rates vary over time, and sometimes bank B’s financing policies change.
For advisors, the orientation consultation is very important as it is the only customer-initiated opportunity to create customer commitment. Normally, when customers choose to take up advice, they also stick with bank B to purchase their mortgage there. So, the main purpose for advisors is to persuade customers to take up advice. In order to do so, they seek to demonstrate a high level of personal service quality. If advisors succeed in presenting themselves as helpful and friendly, as well as competent, polite, reliable and knowledgeable and patiently willing to help customers (Lymperopoulos et al., 2006), the odds are high that customers will choose a mortgage from bank B. Moreover, bank B’s advice costs do not differ much from other financial service providers, so advisors have the freedom to focus only on their personal presentation in order to convince customers to take up advice. Advisor purposes such as these are not explicitly mentioned in the consultations; we derived them from bank B’s documents and checked them with one of bank B’s advisors.
Interestingly, Table 1 immediately reveals a conflict between advisor purposes and customer purposes. The advisor will generally restrict his information to basic mortgage options available to customers: bank B’s policy states that advisors cannot discuss mortgage safeguards other than the national mortgage guarantee (i.e. life insurance, unemployment risk coverage and disability risk coverage). This information is saved for the advice consultation so that customers have a reason to come back. However, customers want to learn about all the decisions they are going to face in the mortgage purchase process, including the safeguards they need to choose.
By investigating the document collection and by analyzing the consultations, we identified the aspect systems and thereupon the aspect system constraints that affect the interaction on the advisors’ side. Many of these non-purpose-related constraints are made explicit in a document called ‘Advice quality and methods’ (only available to bank insiders). The introduction of this document tells us that bank B’s advice quality is based on bank B’s strategy, mission and core values and on the legal requirements as dictated by the Act on Financial Supervision (WFT, 2013). This implies that advisors need to manage and balance various aspect system constraints. Our contextual data led us to distinguish the following six aspect systems that are relevant in the consultations:
the technology aspect system, mainly constituted by the computer program that advisors use in the consultations;
the efficiency aspect system, which provides rules regarding the amount of time available to fulfill the purposes of the consultation;
the legal aspect system, which regulates the design and the amount of information to be provided in consultations as there is a legal requirement to provide customers with correct, clear and non-misleading information (WFT, 2013: Article 4:19);
the bank’s customer service aspect system, which provides various regulations varying from the need to provide a ‘warm welcome’ to customers, to the need to manage the customer’s expectations during the talk, to the rule that no advice may be given in orientation consultations;
the bank’s mortgage acceptance procedure, which sets the criteria to be met for mortgage applicants and determines the customer data that need to be elicited for valid applications;
the interactional pragmatic aspect system of communication, which includes politeness considerations, quality maxims (i.e. providing correct information) and relevance maxims (providing only contributions whose relevance can be reconstructed by the hearer).
Our contextual data suggested two other potentially relevant aspect systems that are not referred to in the consultations:
the internal communication aspect system, which, among other things, is responsible for updating the advisor on changes in mortgage acceptance criteria;
the mortgage application process aspect system; for instance, the final mortgage quotes are not produced by the advisor, but in another business unit.
To the extent that these latter two aspect systems constrain the advisor’s actions, these do not enter in reflexive monitoring because such constraints are typically unknown. For instance, when an advisor has missed a rule update, he is unaware of this.
Collecting design explications
We assembled a collection of 50 design explications, in two ways. First, 10 transcripts were manually screened for references to the constraints suggested by our contextual analysis. Subsequently, a keyword search was done using a list of potentially relevant terms. The keywords included nouns such as time and costs (efficiency aspect system), orientation consultation or advice (the bank’s customer service aspect system); adjectives such as slow or fast (technological aspect system); and verbs such as obliged, may and allowed (legal aspect system). We do not claim that our collection exhausts our data; given the explorative nature of our study, we will not present quantitative findings.
Not every constraint reference constitutes a design explication: sometimes constraint references are ‘stand-alone’ ones, such as in Excerpt 2, line 4:
Excerpts like these are not included in the collection as the technology constraint referred to here (‘with aid of the computer’, line 5) does not conflict with others. So, it does not pose discourse design dilemmas. In contrast, Excerpt 3 demonstrates a constraint conflict:
The current personal data in the computer turn out to be incorrect and require an update (line 1). This need is labeled constraint 1 (a correctness constraint deriving from the bank’s mortgage acceptance procedure, that is, aspect system 5). The advisor explains that the computer program only allows a new address when all the data are reentered (lines 4–5; technology aspect system). We can also infer that satisfying both constraints would considerably delay the consultation; doing things ‘all over again’ (line 5) is clearly undesirable, given the economy aspect system. Her resolution here is to ‘leave it like this’ for now (line 3) and suspend the required correction until after the consultation (lines 1–2). This decision is made explicit so that the customer understands why she does not correct the data, which would be a natural thing to do after checking them. Finally, note that Excerpt 3 shows that more than two constraints may be involved in a constraint conflict.
Considering their sequential environments, it shows that the vast majority of design explications are advisor-initiated. However, there are a few cases; see for instance Excerpt 10 further on, in which the explication is prompted by a customer’s utterance.
Some design explications do not completely list the constraints involved. In those cases, we use our context analysis and the other consultations to reconstruct the conflict; for example, when the advisor mentions a constraint that regularly conflicts with another elsewhere in the data, we assume that the second constraint is also present.
In principle, we may conceive of entirely implicit conflicts, in which no constraint is made explicit at all. For instance, politeness phenomena may be analyzed as attempts to satisfy partially incompatible constraints. But as this article is about explicit discourse design, we will leave those cases aside.
Design explication forms
In our data, design explications take three forms: references to the omission of actions or non-preferred actions (A), accounts (B) and explanations (C). The first two forms may be combined.
Omission and non-preferredness references
Many conflicts are accompanied by references to omissions of actions or non-preferred actions. The difference between these two is a difference of framing. In omission references, the advisor tells the customer s/he will NOT do X as a next action even though X would be desirable; in a non-preferredness reference, the advisor states s/he WILL do Y as a next action even though it violates a constraint. Excerpts 4 and 5 show what these references look like:
In Excerpt 4, line 1, the advisor refers to the need for ‘making acquaintance’, a constraint related to the bank’s customer service aspect system requiring advisors to make customers feel welcome in the orientation consultation’s introduction stage. In the bank’s documents, this is presented as a way to create a bond with customers, which ultimately serves the consultation’s purpose of making customers come back for advice. In line 2, the advisor refers to a second constraint: ‘time’. This leads her to suggest skipping the introduction: line 3 presents an omission reference which is explicitly accounted for; her customer agrees to this in line 9. In lines 10–11, constraint 1 and the chosen resolution are repeated. In order to maintain the focus on customer bonding, she continues by contrastively emphasizing that she does want to hear the customer introduce herself (lines 12–13). Excerpt 5 shows the reference to a non-preferred next action
In lines 1–2, the advisor refers to the computer requirement to enter a fake address in order to continue. This resolution is clearly non-preferred, given that lines 2–3 refer to a common-sense correctness constraint (quality maxim). Lines 4–5 explain why violating this constraint is necessary, by invoking the technology constraint (‘the system needs to know’, line 4).
In both Excerpts 4 and 5, the advisor adds an account for her choice of resolution. In both constraint conflicts, the advisor presents certain constraints as if she has no choice other than to follow them: not following the time constraint in Excerpt 4 and the technology constraint in Excerpt 5 will obstruct the consultation’s progression. Hence other constraints cannot be entirely satisfied.
References to non-preferredness or omission do not necessarily occur with accounts, as we see in Excerpt 6:
The advisor refers to a non-preferred action in lines 3–4: leaving open a field in the computer program that asks for the costs for hiring a real estate agent since the required information is not yet available. The completeness constraint conflicts with the constraint that the advisor can only use information already available.
Accounts
A second indication of constraint conflicts is accounting for the chosen discourse option; we already saw accounts in Excerpts 4 and 5 that are similar in that they use the second constraint to motivate the chosen action. The same goes for Excerpt 3, in which the efficiency constraint is invoked to account for the choice of resolution. Such accounts coupled with omission or non-preferredness references are always provided before or during the advisor action in question.
Accounts may also appear without references to omission or non-preferredness. In our data, stand-alone accounts only occur when advisors reflect on a completed verbal or non-verbal action that helps achieve the consultation’s purpose (implicit constraint 1) but may seem to violate customer expectations. Excerpt 7 shows what this looks like:
Just before the start of this excerpt, the advisor opens an Excel sheet without announcing why. This sheet can be used by customers if they want to make a maximum mortgage loan amount calculation at home and adjust the interest rate periods to see how much they can borrow under what interest rate circumstances. So, it is an extra service to customers (implicit constraint 1, purpose based). However, once the advisor has presented the sheet he realizes that his customers may not understand the relevance of his behavior; this constraint of ‘clarity of relevance’ is referred to in line 1 which prefaces the account.
Explanations
A third form for explications is the use of explanations. Explanations are about legal requirements or the bank’s mortgage acceptance procedure. This focus on legislation and rules designed elsewhere distinguishes them from accounts, which are always concerned with verbal and non-verbal advisor behaviors. Explanations signal that advisors assume that such rule constraints may conflict with customer expectations (hence the second constraint in the conflict is that the advisor’s actions need to follow customer expectations). In such cases, the advisor explicitly refers to the policy or legal constraint to be explained, while the ‘expectation compatibility’ constraint is to be inferred. This is shown in Excerpt 8:
The customer in Excerpt 8 just told the advisor that he intends to go shopping at different mortgage providers for the best offer. The advisor responds by contrasting the nature of the consultation (actually) with the expectation that the customer seems to harbor. Given the new rules as of January 2013, the current consultation is meant for orientation only (lines 2 and 4). In the old days, when these customers purchased their first mortgage (lines 6–10), mortgage offers could be made directly in the first consultation; nowadays, they are made in a second consultation that will need to be paid for (lines 12–14). The upshot is that it will cost the customer a lot of money to shop around for tailored mortgage offers.
In Excerpt 8 the advisor starts explaining the legal changes as of 1 January in response to the customer’s presentation of his reason for coming. In contrast, other explanations (see Excerpt 9) anticipate customer expectations; this kind of anticipation is desirable given the bank’s customer service aspect system.
Reviewing the various explication forms, we may note a difference between explanations and free-standing accounts on the one hand and omission and non-preferredness references on the other. In free-standing accounts and explanations, the customers need to be brought ‘on board’ with the consultation’s design. Once they are, the constraint conflict is eliminated here and now. In contrast, the conflicts underlying omissions and non-preferredness references remain in place once the consultation is over. This leads us to consider the different kinds of conflict resolutions.
Constraint conflict resolution strategies
We have seen that discourse design explications present both the design problem and the solution, that is, some way to resolve the constraint conflict in order to be able to continue their consultation. Three kinds of resolutions can be distinguished:
dropping the losing constraint;
suspending the losing constraint;
integrating both constraints.
AD A
When advisors drop a constraint, they fully comply with the other constraint in the conflict. If we look back at our examples, Excerpt 4 presents a case of dropping, in that the purpose constraint of bonding with the customer (in order to make them return) is not fully satisfied. This compromises the effectiveness of the consultation somewhat.
AD B
Suspending a constraint is postponing its satisfaction. It will be complied with, however, further on in the consultation or once the consultation is finished. Cases in point are found in Excerpts 3, 5 and 6. In Excerpt 3, the advisor suspends correcting the faulty address. In Excerpt 5, she enters a fake address until the actual address will be known. In order not to compromise the advisor’s credibility, these ‘shortcuts’ are presented as technical fixes only, dissociated from the substance of the orientation. Excerpt 6 is also a case of suspending, in that the advisor postpones the satisfaction of the complete information constraint.
AD C
The final option is to find a way of satisfying both constraints: constraint integration. This means that neither of the constraints is dropped or suspended.
Constraint integration is a possible outcome in Excerpts 7 and 8. In both cases, we see attempts to bring the customer’s expectations in line with the advisor’s course of action or the bank’s policies. To the extent that these attempts succeed, this satisfies both constraints at issue. However, to the extent that the customer remains puzzled or unconvinced, the expectation compatibility constraint will need to be given up.
A more complex case involving a combination of suspension and integration is the design explication in Excerpt 9. Here, the customers are expecting a baby and want to buy a house before the woman has given birth. They list a lot of questions and (presumably) expect the advisor to answer them (customer expectations). However, the advisor also needs to fill out bank B’s computer program (consultation purpose constraint). Now, immediately starting up the computer program seems to violate the constraint of being helpful and friendly. Hence the advisor assures the customers that running the program will allow answering their questions along the way:
So, in Excerpt 9 the advisor suspends answering the customer’s questions in order to serve the consultation’s purpose of entering the customer data. In presenting this resolution, she emphasizes that actually there is no conflict between running the program and answering questions, as the program will lead the user through all the important decisions (lines 4–7). She minimizes the effort of running through it all and at the same time discussing important information by using mitigations, such as ‘simply’ (line 2/6), ‘automatically’ (line 3), ‘a few’ (line 7) and ‘just’ (line 12).
To the extent that her account is convincing, she succeeds in actually integrating constraints. Overall, Excerpt 9 confirms the impression from Excerpts 7 and 8 that constraint integration in our data mainly occurs in the context of managing customer expectations. This suggests that expectation compatibility constraints can be satisfied on the spot, that is, by being persuasive, without hurting other constraints; in contrast, other conflicts require a compromise.
Seeking customer acceptance for conflict resolutions
We have shown in various shapes of discourse design explications and have explained how they propose to resolve constraint conflicts. Finally, we will review some presentation strategies that invite customers to accept these resolution proposals. Seeking customer acceptance is a regular feature, as three out of every four discourse design explications is accompanied by one of the strategies outlined below.
Positive framing
Advisors try to ‘balance’ announcements of not doing something by emphasizing what still will be done. In Excerpt 10, the advisor needs to manage customer expectations potentially conflicting with bank policies regarding advice-giving:
The advisor tells the customer that their current talks is an orientation (lines 1–4) and that she will not provide advice about any mortgage risks (lines 6–9). In this excerpt she does not provide an account for that policy, but we know that she has mentioned the legal changes as of 1 January earlier in the consultation. However, after she announces she will not provide advice, which may seem to the customer as if she is unwilling to provide service, she tries to reframe this into something positive by emphasizing all the things she actually is willing to do in the remainder of her consultation (lines 10–12).
We have seen a similar transition in Excerpt 4. The advisor tells the customer that she will skip her own personal introduction (line 10–11), but then she sums up everything she is interested in regarding her customer. So, refusals to comply with customer expectations are regularly followed by some good news that makes the refusal a less categorical one. The positive component is emphasized by its final position in the discourse unit.
Minimizing the problem
The next strategy we identified is downplaying the disadvantages of the conflict resolution, shown in Excerpt 11:
Just before the start of the excerpt, the advisor has experienced troubles with the new mortgage loan calculation program. He uses an older application to resolve this issue. This is clearly a non-preferred option, but it seems the only way to save the consultation’s purpose. Just as we have seen in Excerpt 9, the resolution is presented using a range of adverbials and adjectives that minimize the consequences of the make-do solution.
Requesting customer consent
The last strategy is requesting customer consent, which was already shown in Excerpt 4. Here, the advisor explains that she usually likes to make acquaintances but that time does not allow for this right now. She then asks her customer whether he agrees with skipping it or whether he would like to know ‘the ins and outs’. Consent requests such as these vary in their openness to customer input. Occasionally advisors actually ask customers which constraint they feel should be prioritized, but mostly they clearly project their preferred reaction by using Yes/No interrogatives. Excerpt 12 illustrates how this is done:
In line 7 the advisor suggests suspending his customer’s question as he will automatically get round to answering it later on. In our data, customers never fail to comply with the advisor’s suggested conflict resolutions.
Conclusion and discussion
In this article we have explored discourse design explications: stretches of talk that refer to contextual constraint conflicts and propose resolutions for these. The explications were identified by a combination of top-down and bottom-up strategies. A contextual analysis was used to identify explications, which were then analyzed in terms of form, proposed resolutions and interactional shape. We showed that discourse design explications take the form of references to omitted or non-preferred actions, accounts or explanations. Three strategies are used to resolve constraint conflicts: the advisor drops one of the constraints, suspends one of them or attempts to eliminate the conflict altogether. Finally, we showed how advisors seek customer acceptance of their design proposals by positive framing of their resolution, minimizing the problem or requesting customer consent.
To our knowledge, explicit design explications have not been analyzed in earlierdiscourse-analytical work. Nevertheless, their regular occurrence is interesting from various points of view. First, they are an important exception to the tendency for contextual constraints to remain invisible in interactions; hence they offer a window on how institutional agents navigate the discourse design space. Far from reducing opportunities for ‘agency’, complex constraint sets invite displays of discourse design skills, of which design explications are the most visible specimen. Hence they create new perspectives for the analysis of institutional discourse. Moreover, our method of conceptualizing contextual constraints with reference to purposes and aspect systems constitutes a principled way of linking organizational contexts and interaction analysis, which may be of interest for research into organizational communication.
Of more specific relevance to this particular discourse context is the fact that these displays of advisor skills are at the same time displays of customer centeredness: while providing their consultation services, the advisors are also keen on demonstrating how they optimize these services to suit the customers’ needs. While the agent–customer relation is clearly asymmetrical with regard to knowledge and power, projecting a client-centered image is an important interactional concern of the agents; of course, this furthers the consultation purpose of making the customer return to purchase his mortgage at this particular bank.
Our research has practical implications as well. Investigating discourse design explications highlights the moments when the interaction is under ‘functional strain’, that is, it identifies ways in which the context challenges the participants. While these challenges are primarily addressed in the interactions themselves, the organization may also consider interventions to modify the context: some complexities may need to be addressed by the management instead of by institutional agents and clients. For instance, our explications show that the computer program regularly leads to problems, which clearly invites software improvements. Similarly, the new policies of the Dutch government regarding advice-giving on mortgages require considerable interactional work. Possibly, some of this explanatory work could be moved to other communication media so that the face-to-face contact is not burdened by it.
Having pointed out potential uses of our analysis, we hasten to add that we have certainly not been exhaustive in identifying constraint conflicts in our consultations, as some conflicts will probably not surface in the interaction. First, most customers lack the expertise to fully pursue the purpose of ‘understanding the issues that need to be decided upon when purchasing a mortgage’, as the average citizen is not aware of the various kinds of risk that a mortgage consumer needs to reckon with. As long as the advisor does not bring up these risks, customers will generally not ask for them. Second, customers are not very active in voicing information requests or agenda setting. Hence, our explications mainly concern design problems that immediately threaten the flow of interaction and therefore need to be shared with their customers. In other words, our set of explicated problems is a relatively restricted one. Conceivably, other contexts will provide us with more ambitious forms of explicit discourse design.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.