Verbal interaction in a social dilemma experiment

Full text

Verbal interaction in a social dilemma experiment
Zoe Adams, Agata Ryterska, Devyani Sharma, Magda Osman
Queen Mary University of London
This research was supported by the Social Macroeconomic Hub of Rebuilding
Macroeconomics research network, funded by the Economic and Social Research
Council.
1. INTRODUCTION
Social dilemmas are a fundamental feature of human communication and cooperation. For
example, the 2015 Paris Agreement seeks to slow global warming: ignoring the agreement
would serve a country’s immediate self-interests, but if every country fails to commit to their
pledge, the planet suffers collectively. Social dilemmas of this kind arise constantly, in the
home, workplace and community, and have been studied extensively in psychology,
economics, and philosophy. Although language and communication are central to these
studies, there has been a notable absence of close linguistic analysis of how communication
succeeds in social dilemma negotiations. In this study, we demonstrate a sociolinguistic
approach to the analysis of spoken interaction and propose that this can help illuminate key
factors that underpin success or failure in the outcome of verbal coordination.
The analysis draws on data from a wider study exploring the effect of social forces and
financial incentives on people’s willingness to cooperate (Author et al., in preparation). The
main experiment was a modified public goods game involving 90 participants who were
distributed across 18 groups, six for each of three different conditions. We briefly report on
the methods and main results of the full experiment but then focus our attention on the online
chatroom data of three groups as case studies, one from each experimental condition. In the
design of the experiment, the five participants in each group were given an opportunity to
discuss their initial task performance in an unstructured format before repeating the task.
Through close analysis of these verbal exchanges, we examine how and why certain groups
1

may have been more successful than others in moving towards a consensus for collective
action.
In particular, we import two established concepts from the study of spoken interaction
into the study of economic behaviour. The first is stancetaking—regarded as fundamental to
communication (Jaffe 2009)—which concerns how speakers signal their evaluations of
objects or ideas around them. People can use subtle verbal and non-verbal cues to
communicate stances such as disapproval, authority, or solidarity. These cues have
consequences for (dis)alignment with other individuals, ultimately facilitating or constraining
cooperation. The second construct crucial to interaction is a higher level of sociolinguistic
structure, namely register, i.e., how we speak in different social situations, such as formally or
in slang. Communication in all human societies is organized around frameworks of socially
recognizable registers. Members of a society can use these to adopt powerful ‘voices’ in
conversation, regulated by rules of social convention and constrained agency. In the present
data, we observe speakers employing a range of registers—business speak, game show hosts,
socialist leaders, parental roles—in order to align with or disalign from each other, and to
assert their preferred stances. A close examination uncovers why these verbal acts might
succeed under some conditions and not others.
We might ask why a classic behavioural economics experimental design should warrant a
sociolinguistic analysis of verbal interaction. The field of economics is grounded in a
quantitative and mathematical foundation that tends to steer clear of more qualitative
questions that may seem too subjective or intractable (Lenger 2019). Sociolinguistics as a
discipline shares this positivist epistemology but, as its object of study is spoken interaction,
it integrates quantitative generalization with qualitative detail. When behavioural experiments
involve spoken interaction, this additional qualitative dimension of how language organises
inter-personal exchange and persuasion becomes very relevant. Conventional quantitative
2

data allow researchers to generalize over a large sample, in this case pertaining to group
consensus outcomes, but a mixed qualitative and quantitative examination of the verbal
interaction itself complements this with an understanding of what is occurring incrementally
and in real-time within groups to achieve those outcomes.
In the 1960s, the noted sociologist Erving Goffman—a founding figure for interactional
sociolinguistics—exchanged ideas on the study of social games with economist Thomas
Schelling at Harvard. While sharing economists’ interests in the formal properties of games,
Goffman (1961) observed that any study of social encounters must additionally take into
account social and psychological dynamics that often lack sufficient consideration in formal
approaches to game theory. He remarked, “a game move is one thing; self-mobilization
through which this move is executed during a gaming encounter is quite another. Game rules
govern the one, the structure of gaming encounters governs the others” (p. 38).
Despite these fruitful early exchanges, the interdisciplinary interface between purely
rational calculations and the social framing available to agents for their resulting actions
remains under-examined. We have seen extensive adoption of game theory and rational
choice theory in linguistics in recent years (Burnett 2019; Franke 2009; Goodman & Frank
2016; Jäger 2008; Myers-Scotton & Bolonyai 2001), but almost no application of
sociolinguistics in economics. With this case study of verbal encounters in the course of a
social dilemma, we revisit the promising terrain that Goffman and Schelling explored over
half a century ago.
2. BACKGROUND
2.1 Communication in social dilemmas
In social dilemmas, the rational strategy is to defect, so the high degree of cooperation
observed in experiments has led to a substantial body of work aiming to understand people’s
3

motivations. Many of these studies have confirmed the important role of language. A meta-
analysis of 45 studies revealed that communication greatly increased cooperation (Cohen’s d
= 1.01) (Balliet 2010). Researchers have also found that this effect extends to written
communication in a chatroom despite the absence of rich nonverbal cues (Zheng et al. 2002).
A narrower body of research has explored how exactly group discussions affect
cooperation. One account proposes that communication induces a norm of cooperation within
participants who then believe that prosocial actions are expected (Bicchieri 2002). Another
draws on Tajfel and Turner (1979), arguing that communication fosters a common bond,
which creates an enhanced group identity (e.g., Orbell et al. 1988). In turn, members view
themselves as part of the same in-group and are more inclined to cooperate. A final account
argues that communication provides the opportunity for promise-making (Bouas & Komorita
1996; Kerr & Kaufman-Gilliland 1994). In order to avoid cognitive dissonance, individuals
aim to uphold their public commitments. In this case, cooperation results due to a personal
norm rather than a social norm.
These approaches invoke largely social psychological accounts of the role of
communication, which can leave unexplained why some instances of communication are
more successful than others. For example, Zheng et al. (2002) adopt a common research
design that examines different communication conditions (face-to-face, social chat, photo,
personal details, no communication) but not the talk itself. A number of economists have
therefore gone further and performed content analyses of discussions during social dilemmas
(e.g., Goren & Bornstein 2000; Kagel 2018; McClung et al. 2017). Much of this work is also
informed more by psychology than linguistic theory, and so stops short of analysing
persuasion through unscripted talk. Others have explored the role of language from a
pragmatic perspective with an emphasis on Searle’s (1969) notion of speech acts (Sally 2005;
Shank et al. 2019).
4

We add established sociolinguistic tools for the analysis of talk to this body of work.
Building on the work of Goffman, Sally (2000, 2005) makes a case for this socially informed
perspective in the study of social dilemmas alongside the established notion of cheap talk:
“talk is not cheap, rather it is complicated, creative, implied, affective and
effective, altering speaker and listener and any embedded game […] once you
let in a drip or two of literal meaning, you cannot prevent all of language from
flooding in: poetry, promising, metaphor, irony, insult, intimacy, and the rest.”
(Sally 2005, p. 263)
This kind of acknowledgement of the complex reality of spoken interaction is still rare in
behavioural economics. We argue that this “complicated, creative, implied, affective” flood of
language is in fact a tightly structured signalling system that plays an important role in
determining the extent to which cooperation is achievable in a given setting.
2.2 Stancetaking in communication
We adopt Du Bois’s (2007) widely used model of stancetaking in this article. Du Bois defines
stance as “a public act by a social actor, achieved dialogically through overt communicative
means (language, gesture, and other symbolic forms)” (p. 163). Speakers typically develop
stances in response to those taken by other interlocutors rather than by a high-level prescribed
norm. Norms for an interaction therefore develop dynamically and ‘intersubjectively’—in
response to preceding acts by others—within the context of wider, established rules for
engagement. Du Bois (2007, p. 173) argues that stancetaking, the “smallest unit of social
action”, is always implicated in an utterance, as any choice of words reveals a stance, e.g., the
5

choice of alleged over said. His model combines two core concepts: evaluation and
alignment.
Evaluation is the assignment of value to an object, either a referent external to
discourse or an element of the discourse structure itself. For example, if a person says what a
great holiday!, they are positively evaluating a discourse object external to the interaction,
but if they say obviously, I know that, they are negatively evaluating the previous speaker’s
utterance or turn, an object within the conversational structure. In this way, we can evaluate
anything, but from a social dilemma perspective, stance objects might include the reward
(external to discourse) or a previous speaker’s suggested strategy (discourse itself).
In evaluating an object, speakers position themselves along both epistemic and
affective scales. For instance, I know is an epistemic evaluation that focuses on how certain
an individual is about their assertion, whereas I’m glad is an affective evaluation, an
expression of their emotional relation to the speaker or to a discourse object. Both scales are
important in interaction. Jacknick and Avni (2016) argue that epistemic stancetaking is
pivotal in anonymous scenarios particularly around high-stake topics. A social dilemma is
such a scenario; indeed, our experiment entailed both financial and social risk. Defectors
could win up to £70 if their entire group cooperated, but this would potentially expose them
as a selfish freerider. Cooperators could win far less if everyone else defected, but this would
frame them as altruistic. Epistemic stancetaking in such situations allows an individual to
establish credibility and assert expertise. Regarding emotion, neuroeconomic research attests
to its important role in decision-making, for example irrational decision-making resulting
from aversive affect (Engelmann & Fehr 2017). As we will see, the present scenarios give
rise to a range of intense emotions, ranging from anger and shock to relief and delight.
The second component of Du Bois’s model is alignment. Any act of evaluation is
typically also an act of alignment with (or disalignment from) other actors in the
6

conversation. Figure 1 presents Du Bois’s visualisation: As individuals evaluate objects, these
evaluations position them relative to other evaluations in the discourse, which implicates
inter-subjective alignment.
Figure 1. The Stance Triangle (adapted from Du Bois 2007, p. 163)
Alignment is the focus of our analysis in the present paper. It is useful to distinguish between
two aspects of alignment: evaluation and structural (see Kiesling et al. 2017 for further
dimensions). Evaluation alignment indicates agreement with a speaker on their evaluation of
a given object; this is paramount in a social dilemma, as cooperating is only better than
defecting if everyone agrees on a strategy. However, structural alignment, which involves
faithfully participating in the structure of an activity, such as providing an answer to a
question or formatting a response using the same syntax as the prior speaker, is equally
significant. If an individual asks whether everyone agrees to cooperate, and one member of
the group avoids answering, this can damage group trust. In considering both structural and
evaluation alignment, we can observe how a speaker’s original stance comes to be endorsed,
transformed or undermined by the next speaker (Jaffe 2009).
7
SUBJECT 1
SUBJECT 2
STANCE
OBJECT
e
va
luat
es
e
va
lua
t
es
aligns
posit
ions
posit
ions

It can be difficult to recognise whether or when a group has reached a decision from
an organisational perspective because the process entails “incremental activities consisting of
many minor steps” (Halvorsen 2015, p. 2). Participants review their understandings of the
communicative event after each utterance, and any stancetaking analysis must respect that
consensus is achieved at this micro level (Kärkkäinen 2006). To track these increments
systematically, we follow Kiesling et al. (2017), who attempt to overcome the intrinsic
subjectivity of interactional research by embedding Du Bois’s categories within a
computational implementation for measuring stancetaking, with corpus annotation and
analysis. We additionally devise a format for visually tracking consensus formation via
diverse stancetaking moves as they unfold among participants. We also follow Kiesling et
al.’s use of a scale of 1-5 to code online posts for stance elements. Their nuanced treatment of
alignment as continuous rather than dichotomous helps to show how participants maintain
“strategic ambiguity” (Du Bois & Kärkkäinen 2012, p. 440)—a crucial device in a social
dilemma game, given the trust and risks involved.
2.3 Frame and register in communication
As soon as an actor resolves to adopt a given stance during an interaction, they are faced with
the problem of how to signal it. Stances are not rigidly tied to specific linguistic formulations.
A person might express positive affect with I’m delighted or yippee! or any number of other
alternatives. In each case, their choice of linguistic signal brings with it a conventionalized
frame or “schema of interpretation” (Goffman 1974), a socially shared set of principles for
classifying and interpreting experience. The expression yippee! may invoke a playful frame,
implicating childlike participant roles and reduced social distance. If a participant succeeds in
imposing a given frame, it can limit other frames in the interaction and bind others to the
appropriate behaviours for that frame.
8

Our linguistic choices thus involve the execution of recognisable social registers of
speech (Agha 2005; Bakhtin 1981; Biber & Finegan 1994). A register is a particular type of
language associated with specific social groups or speaker roles, for example parents, news
reporters, medical professionals, gangsters, and innumerable other culturally
conventionalised ways of being. Du Bois (2011) observes that analysing language does not
just require a focus on the present moment of interaction between actors (stancetaking), but
also exposure to discourses of the past within actors (register). The expression of stance thus
relies on registers circulating in the wider social context (Jaffe 2009, p. 4). Words carry the
traces of their past usage, “the social atmosphere of the word” (Bakhtin 1981, p. 277). Once
an individual introduces these other “shadow” authors through their stylistic choices, their
chosen stance is no longer theirs alone; the invoked ally—a parent, a confidante, a political
leader—can consolidate the force of the stance (Coreen & Sandler 2014). These
“sociohistorical snapshots” (Agha 2015, p. 27) become powerful resources for stancetaking
by speakers as well as interpretation by hearers.
When we analyse stancetaking, then, we must additionally take account of the articulation
of stances through voices, or recognisable personae or figures, in conversation. Even when
actors believe they are simply being themselves, they are typically using a particular voice
and framing, suited to the situation. Within every person lies a directory of different ways of
speaking, and we use these voices as a tool for persuasion.
To date, economics research on social dilemmas has not examined how these universals
of stancetaking and register lead actors to achieve or change their interactional goals. In
identifying the main arguments for high levels of cooperation, Kagel (2018) includes the
following example from one participant: “yo other team if you trust us we can both choose a
(cooperate) and make some hashtag cash.” The use of slang, pop culture, and social media
lexicon deploys an informal register that could have been an attempt to reframe ingroup-
9

outgroup boundaries and elicit cooperative behaviour. Its uptake would have depended on
ambient stances and registers asserted by the other participants. These acts of conversational
framing were not analysed but likely played a part in the outcome.
We examine the communicative strategies of groups that succeeded or failed to reach
consensus. As group members use a wide range of discourse strategies to achieve
interactional goals in professional exchanges, from humour to authoritative language (cf.
Koester 2010), we examine all salient strategies.
3. METHOD
3.1 The social dilemma experiment
The data for this paper come from a larger study exploring the impact of reputational
information on decision-making in social dilemmas. The study is adapted from the “repeat-
play” public goods game (e.g., Muehlbacher & Kirchler 2009) which comprises two rounds
with a Task and Chat element as laid out in Figure 2. 90 participants (11 males, 79 females)
were recruited using opportunity sampling. They were all aged over 18 with no conditions
that limited the use of their non-dominant hand. Participants were divided into groups of five,
based on the order in which they were recruited. There were 18 groups, six groups for each of
the three conditions.
After being assigned to a group with four others, participants completed the Effort
Task which involved squeezing a hand-grip device 40 times. As this requires effort, each
squeeze was associated with a 40p reward. The reward resulting from each squeeze (i.e.,
effort) could either go to an individual pot or a group pot. Rewards from the individual pot
were given directly to the participant. The contributions that the participants made to the
group pot were multiplied by 1.5 and divided evenly among all members of the group. A
10

participant’s final reward therefore depended on how much effort they contributed to their
individual pot, as well as on how much all group members contributed to the group pot. Prior
to squeezing the device (Execution Stage), they were asked to state how many squeezes they
intended to contribute to the group pot (Choice Stage). For the Online Chat, approximately 2-
6 days later, participants joined a platform called E-chat where the experimenter revealed the
results. They had the opportunity to discuss their performance over a period of 36 hours. This
stage involved the crucial reputation manipulation where the experimenter revealed three
types of information depending on condition: (1) intention – how many squeezes each
participant intended to contribute to the group pot, and the group pot total; (2) action – how
many squeezes each participant actually contributed to the group pot, and the group pot total;
(3) baseline – the group pot total. Participants were required to post at least one task-related
comment. In Round 2, they repeated the Effort Task on average 4.8 days later, and the
experimenters ensured that they had read the online discussion before doing so. They then
took part in another Online Chat. Participants could earn up to £70 ($94), but, for ethical
purposes, all participants received a minimum of £40 ($52) to avoid vast discrepancies in
payment. The wider study tested hypotheses relating to feedback, reputation, and cooperation
(Author et al., in preparation).
The present study briefly reports on the overall findings but then focuses on the
Online Chat of Round 1, and in places Round 2, of three groups, one from each condition and
covering a range of behavioural outcomes, from defection to full cooperation. The three
discussions ranged in length from 13-15 comments.
11

Figure 2. Schematic of experiment
3.2 Coding and annotation
For the coding of alignment, we closely follow Kiesling et al.’s (2017) process, which begins
with identifying the object of evaluation, or “stance focus”, for each utterance in each
discussion. If a comment included multiple utterances by a participant but no single stance
focus for the entire comment, then the comment was split according to stance focus, as in (1).
(Subject IDs take the form of group name, subject number, gender.) Equally, if there were
multiple consecutive comments with the same stance focus, then the comments were joined
as in (2). This ensured that each annotation was linked to one stance focus. Greetings were
excluded or, if followed by another comment, they were merged as in (3).
(1)Green5F (22:49): This looks a really interesting study! (Focus: study) I put in all of my
40 in the group pot in the first round and will do the same in the second round
(Focus: contributing all 40 to the group)
(2) Gold5M (09:11) Hi everyone, that's a good amount! The more we contribute the more we
will get since it's x1.5 (Focus: Round 2 strategy)
12

Gold5M (09:12) Shall we go all in for part 3? If we all do, instead of £16 we each get
£24 (Focus: Round 2 strategy)
(3) Green2M (09:08): Hey guys! thats a decent group score, lets use it as motivation to get a
bigger pot (Focus: Round 2 strategy)
Each utterance was then annotated for a value of 1-5 for alignment (the same number of
distinctions used in Kiesling et al. 2017). A score of 5 indicates high alignment, 1 indicates
low alignment. The first and third author double-coded the first discussion blind and
compared their results (Cohen’s kappa for inter-rater reliability: 0.654); disagreements were
examined to develop more explicit heuristics for ambiguous cases. The first author then
proceeded with coding discussions two and three. Kiesling et al. (2017), who used a large
number of coders, observe that one reason for disagreements in their annotation process was
that each conversational thread introduced new challenges due to a wide variety of topics,
ranging from parenting to fitness and sometimes requiring inside knowledge. The discussions
in the present analysis, by contrast, are narrowly focused and all centre on the social
dilemma. For this reason, despite similar data, we had a reasonably high inter-rater reliability
score.
Kiesling et al. (2017) additionally coded for investment (level of commitment to the
stance focus) and affect (evaluative polarity of stance towards the stance focus). We
conducted a preliminary examination of all three. In many cases, these further factors pattern
with the alignment measures; however, for the specific question of how consensus towards a
specific strategy is achieved, and given the narrow thematic range of exchanges, we restrict
our focus to alignment as the most pivotal dimension for the present analysis.
13

Table 1 displays examples of utterances that received an alignment score of 1 and 5.
The utterances are listed along with the preceding comment, which helps to clarify the
alignment scores. Although we do not analyse evaluative polarity here, the stance focus in the
final column shows how alignment emerges in part through evaluation of stance foci.
Table 1. Examples of alignment coding
Ex
.Username Comment Alignmen
t
Stance focus
1 Gold4M we can do around 25-30 in the pot Strategy for Round 2
Gold2M why not maximize it and put all 40
in the pot? bigger the contribution
bigger the reward don't you think?
1 Gold4M’s suggestion on
strategy for Round 2
2 Green4F I’m happy to add all credits to the
group if everyone is in agreement
Strategy for Round 2
Green1M Yay!! 5 Green4F’s promise to
cooperate
Kiesling et al. (2017) did not annotate the first posts for alignment in their data, as they were
initial in the interaction. By contrast, in the present data the first post was always the
experimenter, who revealed the results from the task in Round 1. As the experimenter’s
comment always implicated the entire group’s behaviour, the following comment—the first
by an experimental subject—was always aligning or disaligning to some degree with the
group. For example, in (4), we can see that Gold1F, who commented first, is aligning with her
group by positively evaluating the performance of all members, albeit to a slightly weak
extent as suggested by the qualification of “too”.
(4) Gold1F (09:00): Everyone’s contributions are not too low! Which is a good thing.
We cannot straightforwardly generalise the findings that we present for three case studies to
all other groups. However, we do start with a brief summary below that shows that
consensus-building discourse in our data does correlate overall with an increase in collective
14

behaviour. This complicates proposals that assert that communication per se increases
cooperation, for social or psychological reasons. Under that view, we should not expect to see
a correspondence between the quality of the interaction and the level of success in a
collective action outcome.
4. ANALYSIS
As a first step, we explored the effect of online discussions on levels of cooperation for all 18
groups. This provides background information for our closer analysis of within-group
interaction. A number of independent variables were examined; the key independent variable
for the present discussion was the consensus score for each group.
In the wider study, we used an aggregate score for consensus talk that was distinct
from the alignment analysis presented in the next section. For this consensus score, we
identified all consensus-building speech acts (e.g., encouragement and promises to cooperate)
in each transcript and assigned each comment of this type a value between -1 to 1, in
increments of 0.25, as a measure of how much it supported a group consensus to cooperate.
Inter-rater reliability was tested via blind double coding of 5% of the data: Agreement within
1 level out of 9 was at 75% before discussion and 85% after. Each group received an average
‘consensus score’ which served as an independent variable alongside other factors in a linear
regression analysis, using the lm function in R (Bates et al. 2015). Table 2 shows the
difference in group squeezes between Round 1 and 2 and consensus score for each group
which is visualised in Figure 3. In order to test for an effect of communication on levels of
cooperative behaviour, the difference in group squeezes was taken as the dependent variable,
whereby a higher score indicated a greater degree of cooperation. Results revealed a
significant effect of consensus score on cooperation ( = 92.54, F(1,16) = 11.98, p = .003).
15

Table 2. Consensus score and the difference in group squeezes between Round 1 and Round 2 for
each group
Group Differences in squeezes Consensus score
Mauve 115 0.696
Green 85 0.769
Yellow 82 0.675
Rose 77 0.367
Blue 70 0.800
Purple 65 0.350
Navy 57 0.667
Orange 53 0.528
Violet 46 0.600
White 44 0.357
Grey 43 0.500
Gold 42 0.734
Plum 34 0.306
Red 32 0.292
Silver 30 0.050
Black 20.500
Turquoise -15 0.286
Scarlet -16 -0.039
16

Figure 3. Consensus score and the difference in squeezes between Round 1 and Round 2 for all 18
groups
This wider analysis addressed the question of whether degree of consensus within
communication impacts upon later behaviour. The result corroborates previous claims that
not only does communication have an effect on cooperative behaviour, but the quality of
communication is also operative (e.g., Charness & Dufwenberg 2006; Chen & Komorita
1994; Dawes et al. 1977; Kagel 2018; Oprea et al. 2014; Ostrom et al. 1992; Zheng et al.
2002).
The finding that some discussions were more successful than others at generating more
cooperative behaviour leaves a further question unanswered: why did some discussions move
towards more cooperative talk than others? In the case studies that follow, we conduct a close
analysis of the strategies participants use to build consensus through talk, and why they meet
with mixed success.
One group from each condition, marked in bold in Table 2, was selected for a mixed
methods analysis combining qualitative and quantitative tracking of discourse: Orange
17

(action), Gold (intention), and Green (baseline). These were selected because they capture a
range of degrees of success in achieving collective behaviour, from defection to full
cooperation. This range provides a rich array of interactional styles for stancetaking and
register analysis to examine. Given the small scale of the data, we refrain from claiming any
direct causality between observed interactions and subsequent behaviour; instead, we refer
the reader to the quantitative result above, showing an overall significant effect of emergent
consensus on behaviour.
For each group, we first present a summary of how many squeezes each participant
contributed to the group pot out of 40 in the Effort Task of Rounds 1 and 2 (Tables 3-5), and
then analyse the Online Chat in Round 1. This is presented alongside an alignment graph
tracking the strength of alignment of each comment in relation to the previous participant’s
comment. This delves into the incremental shifts in interpersonal alignment that partly
underpin the composite consensus scores in Table 2. The transcripts exclude the first post, in
which the experimenter reveals the results, as the format of those posts always comprised a
greeting to the group, and then the following: Condition 1 (intention): “Username[1-n] chose
to contribute X squeezes to the group pot” and the group win; Condition 2 (action):
“Username[1-n] contributed X squeezes to the group pot” and the group win; Condition 3
(baseline) “The group win for Round 1 is £X”.
We begin with the group that was most successful in pursuing collective action in Round
2, followed by groups that had increasingly unsuccessful outcomes for collective behaviour.
4.1 Green team: The family gameshow
Table 3 shows that the Green team (baseline) was extremely successful in increasing their
cooperative behaviour following their discussion. After the Online Chat, every participant
increased their contributions to the group pot to the maximum number of squeezes.
18

Table 3. Squeezes contributed to the group pot by Green team participants in Round 1 and Round 2
User Round 1
Discussion
Round 2
Green1M 40 40
Green2M 0 40
Green3F 15 40
Green4F 20 40
Green5F 40 40
Total 115 200
Figure 4. Transcript of discussion for group Green. Sequential comments with the same stance focus
by a single participant are assigned one alignment score. Comments separated by ‘//’ have a different
stance focus and are coded separately.
19

Figure 5. Alignment score for each comment in relation to prior speaker in Green team discussion (1
= disaligned, 5 = aligned)
Figure 4 provides the transcript of this group’s discussion and Figure 5 provides the
alignment visualisation. Our analysis centres around the interactional frame created by the
first commenter, Green1M, and the resulting tension and resolution in the chat.
In his opening comment, Green1M first reduces social distance with a playful
solidarity marker (“Peeps!”), then delivers a reputational threat. The use of morally
evaluative terms such as “altruistic” and “greedy” without further justification and little
verbal hedging construct an authoritative role (van Leeuwen 2007), even before the result is
announced. Green1M maintains this paternalistic role with the framing device “scores on the
doors”, popularised by the host of the television show The Generation Game, an inter-
generational family contest. He embroiders these framing devices with ambiguously playful
signals: double exclamation marks intensify the emotive force of the moral sanction (Ip 2002)
but can also seem comically hyperbolic; the emoticon hints at his own vulnerability for
trusting his group but also issues a warning to others of the big reveal of their moral compass.
20

In this way, he comes across as the family judge presiding over everyone’s actions (Ochs &
Taylor 1995).
This compact but powerful package of signals invokes a domestic disciplinary frame
and potentially constrains responses available to others. Recall the link between structural
(dis)alignment and evaluation of discourse objects: if participants risk disaligning from
Green1M structurally by failing to acknowledge his comment, Green1M’s evaluation—in
favour of the collective strategy—might also have been less valued. As it turns out, Green1M
is wholly successful in bringing the group around to his evaluation. How does he achieve
this?
The first factor is that his main ‘rival’, Green2M, adopts a conciliatory strategy.
Green2M is the only player to contribute 0 and so is a direct target of Green1M’s threat
(though only he is aware of this). In comment 2, he avoids the risk of losing face by opting
for a positive tone and placating Green1M: he praises the group result, echoes Green1M’s
gameshow term “score” (not used in any of the other 18 groups), and uses the imperative
“let’s” to share in Green1M’s authority.
Instead of aligning with Green2M, Green1M escalates the face threat to Green2M and
the group by conveying the full force of personal injury with “Ouch.” As the full stop is
optional in texting, its presence can express subtle interpersonal information (Houghton et al.
2018), in this case dismay through a move away from his high-affect exclamation marks. The
gameshow-esque “£69 out of a possible £120” highlights the magnitude of his
disappointment. From an economic perspective, these explicit signals of disappointment are
an act of social shunning, which function to offset the perceived short-term benefits of
uncooperative behaviour and reduce the incentive of freeriding (Gachter & Fehr 1999). The
parental tone (imperatives, exhorting good behaviour, lack of mitigating politeness)
reintroduces the original status imbalance and positions the group’s problematic behaviour as
21

the focus of his negative evaluation. Green2M again repairs potential damage in comment 7
by seizing the opportunity to rehabilitate himself positionally and aligning explicitly with
Green1M.
A second overall factor facilitating Green1M’s success is his artful double-voicing of
authority (parental/gameshow frame) and solidarity (socialist frame), softening the
unmitigated format of the former stances with humour and positive affect in the latter. As
support for his stance starts to be voiced, Green1M quickly abandons his risky stern register
and adopts almost the reverse register, a childlike “Yay!!” followed by a collectivist call to
arms: “All for one and one for all!!...£££!!!”, with a return of an excitable tenor through
punctuation. Even his vacillations in mood are in line with the emotive nature of gameshows
which “excite passions” as contestants “weep and dance, kiss and hug, scream and shout”
(Roe et al. 1996, p. 49). More than a jumble of stances, the combination of Green1M’s
different moods can be viewed as a single performance. This allows us to see that, despite the
risks involved, such displays of emotion can spark social action (Du Bois & Kärkkäinen
2012).
The final factor is Green5F’s closing confirmation. Her positive evaluation with an
asserted stance object of “everyone getting on board” implicates the entire group, including
Green3F who structurally disaligned from Green1M in comments 5-6. She places on record
her full cooperation in Round 1, exempting herself from Green1M’s earlier punishment, and
structurally mirrors Green1M’s animated gameshow affect with a smiley emoji, exclamation
marks, his phrase “come on”, the game term “round” (only featured two other times in the
entire dataset), and the communal phrase “band together” (Twenge et al. 2012).
Comments made in the Online Chat of Round 2 are also revealing. Drawing on
Goffman (1956) and Brown and Levinson (1987), economists Dunning et al. (2014, p. 124)
argue that participants in social dilemmas cooperate because it is insulting to withhold trust.
22

In other words, they honour an injunctive norm to trust “not because it is what they want to
do but because they feel it is an obligation of their current social role.” When the
experimenter revealed that everyone cooperated in Round 2, Green4F said: “Wow great job
everyone, no one betrayed the group !” While claims of causality cannot be made, it is
interesting to observe Green4F’s cooperative behaviour in light of Green1M’s parental stance
and her astonishment in Round 2. Her surprise suggests that she trusted her group at the
behavioural level but not at the cognitive level, along the lines of Dunning et al.’s proposal.
Taken together, despite a rocky beginning, the discussion ends with a high degree of
alignment and full cooperation in Round 2. We see how an agent can make a chosen norm
salient to induce cooperative behaviour (Bicchieri and Lev-On 2007), leading to subtle
structural and evaluative uptake in the discussion. Green1M created a group identity (Orbell
et al. 1988) through registers associated with cooperation and the discussion shows how trust
can be fostered at zero acquaintance for the greater good (Dunning et al. 2014). As Du Bois
(2007, p. 173) observes: “we care about the state of the game, too: how it is played, who
plays it well and fairly, in what condition the players leave the turf.”
4.2 Orange team: The politician
Table 4 shows that Orange team (action) was not very successful in increasing their
cooperative behaviour following their discussion. Although the number of squeezes
contributed to the group pot increased, only two participants contributed the maximum
amount (Orange1F, Orange3F), only one changed their behaviour in this direction, and two
failed to increase their contributions at all (Orange4F, Orange5F). As we will see, this group
is less successful in aligning during their interaction as well.
Table 4. Squeezes contributed to the group pot by group Orange participants in Round 1 and Round 2
User Round 1 Round 3
23

Discussion
Orange1F 15 40
Orange2M 0 29
Orange3F 40 40
Orange4F 20 19
Orange5F 4 4
Total 79 132
Figure 6. Transcript of discussion for team Orange. Sequential comments with the same stance focus
by a single participant are assigned one alignment score. Comments separated by ‘//’ have a different
stance focus and are coded separately.
24

Figure 7. Alignment score for each comment in relation to prior speaker in team Orange discussion (1
= disaligned, 5 = aligned)
Figure 6 provides the transcript of this group’s discussion, and Figure 7 provides the
alignment visualisation. In Figure 7, we see a steadily declining level of alignment, which
does not bode well for an outcome of increased collective behaviour.
In the Orange team, the core difficulty appears to be an unresolved tension in register
choice (potentially arising out of underlying stance differences) and lack of harmonisation
between Orange3F and Orange4F, which impedes trust and coordination overall.
The discussion begins with Orange1F and Orange2M agreeing to increase their
contributions. The alignment slope in Figure 7 starts to decline when Orange3F tries to
encourage full cooperation and Orange4F expresses distrust. This triggers a power play
between the two. Unfortunately, despite a highly agreeable comment by Orange1F in
comment 11, the discussion never recovers and Orange5F’s final non-committal promise
leaves the discussion on an uncertain note.
Let us begin with the political tone of Orange3F. Although Orange1F and Orange2M
did not greet the group, Orange3F joins the discussion with a formal greeting in comment 4
25

immediately followed by a disaligned stance in favour of full cooperation. She draws
attention to her cooperative behaviour in Round 1, emphasising others’ lack of altruism
through her use of “you”. A number of features of her discourse channel political speech. The
present perfect “have given” operates as a persuasive technique in the realm of politics,
linking past decisions to the present moment (Fetzer & Bull 2012), as does the strong
epistemic certainty of “I believe” (Fetzer 2014). This contrasts with the lower epistemic
commitment in earlier comments, in forms such as “I think”, “may”, and other hedging
devices (Milkovich & Sitarica 2017). Orange3F uses the inclusive pronoun in “we should” to
momentarily align with her group and present her goals as the audience’s goals, a persuasive
technique used by the “inspiring orator” of politics (Joseph 2006, p. 13). However, when
subsequently eliciting her group’s thoughts, she does not opt for a low social-distance format
such as ‘What do you think?’, but rather asks if anyone shares her opinion, positioning it as
the norm. The clause that follows (“Logically…”) again strongly asserts epistemic authority
and discourages disagreement. The formal, authoritative, and somewhat argumentative nature
of her comment is typical of political discourse (Archakis & Tsakona 2010; van der Valk
2003). Framing her comment as a speech, she plunges the group into the role of an audience
(Jaffe 2009). As with Green1M in the first case study, Orange3F places herself in a risky
position with this bold framing choice, positioning her as an authority, but somewhat lacking
the nuance of Green1M’s frame-shifts.
Although Orange3F’s register ultimately fails to unite, it influences the format of later
comments. Orange2M aligns with Orange3F in a tone of equal formality. The content of his
comment, however, is not entirely aligned: the subtle inclusion of “if we all do it” reiterates
the risk. This is reinforced by Orange4F who first disaligns from Orange2M by labelling the
results as “interesting” and offering a slightly critical assessment of Orange3F’s generosity.
Orange4F maintains an uncertain stance to Orange3F’s proposal, explicitly introducing
26

concerns over trust. In this unsettled, low alignment mood, Orange3F attempts to regain
control (comments 9 and 10) with a moral argument to invoke cooperation by making an
emotional appeal to their “human decency”, “hope”, and “cooperation”, still in the political
oratorial register (Duranti 2006) but with a note of frustration in the punctuation. Her use of
netspeak “u” and awkward, possibly non-native use of “retaliate” is somewhat incongruent
with her chosen register, risking her credibility. Goffman (1956, p. 33) remarks our tendency
to “pounce on trifling flaws as a sign that the whole show is false’ and this ‘forces an acutely
embarrassing wedge between the official projection and reality.”
In a final blow to Orange3F’s authority, Orange4F upends Orange3F’s imposed frame
with a phrase popularised by a hip-hop album in the 2000s entitled Get Rich or Die Tryin’.
Like Green1M’s direct quotation from a gameshow, the reference here is unambiguous, in
particular its stark contrast to the political register in play: hip-hop symbolises resistance
against status-quo politics (Perry 2004). Her surface sentiment is positive, but the structural
disalignment, including lack of direct relevance, from Orange3F is painfully evident.
Orange3F is attempting to resist her mistrust, but Orange4F responds with mockery and
defiance. By contrast, Orange1F does perform a “stance follow” (Du Bois 2007, p. 161), i.e.,
takes up the action made relevant by Orange3F, agreeing to contribute all 40 squeezes to the
group and executing this promise in Round 2. This is the only moment in the discussion
where alignment, evaluation and affect are all high. The conversation ends with Orange5F
agreeing to cooperate, swiftly followed by a reinterpretation of Orange3F’s strategy as
“contributing at least something”. Indeed, Orange5F contributed only 4 squeezes in Round 2,
as in Round 1.
In this group, Orange4F and Orange5F failed to cooperate. Bicchieri and Lev-On
(2007) argue that in the impoverished online chatroom environment, without auditory or
visual cues, participants are more removed from settings where promises are made and kept.
27

Furthermore, this group was in the action condition, and so their low contributions in Round
1 are visibly inconsistent with their pledges to cooperate, raising legitimate concerns about
defection (Wilson & Sell 1997; see also Gold team in 4.3 below). These trust issues are
exacerbated by Orange3F’s awkward attempt to impose her stance through a political voice,
with resistance from Orange4F. Disagreement in groups immobilises collective action
(Goffman 1956). In fact, studies have found that discussions of mistrust are absent in
cooperative groups (Goren & Bornstein 2000). This is line with Majeski and Fricks (1995, p.
629) who speculate at the outset of their paper: “saying the wrong thing in the wrong way
(e.g., saying something that the opposition interprets as insincerity or untrustworthiness) can
be worse than not talking at all and might harden group-based distrust”. Such an
interpretation receives support in Round 2 when, upon discovering the cooperative behaviour
of her group, Orange4F remarks: “wow, I guess I need to be more trusting”.
4.3 Gold team: The business meeting
Gold team (intention) achieved only limited overall success in increasing their cooperative
behaviour following their discussion. Table 5 shows three participants did shift fully to
cooperative behaviour in Round 2, but two defected (Gold1F, Gold5M). In this final case, we
highlight some limitations of chat data, particularly due to sincerity violations.
Table 5. Squeezes contributed to the group pot by Gold team participants in Round 1 and Round 2
User Round 1
Discussion
Round 2
Gold1F 12 5
Gold2M 15 40
Gold3F 20 40
Gold4M 6 40
Gold5M 30 0
Total 83 125
28

Figure 8. Transcript of discussion for group Gold. Sequential comments with the same stance focus
by a single participant are assigned one alignment score. Comments separated by ‘//’ have a different
stance focus and are coded separately.
Figure 9. Alignment score for each comment in relation to prior speaker in Gold team discussion (1 =
disaligned, 5 = aligned)
29

Figure 8 provides the transcript of this group’s discussion, and Figure 9 presents the
alignment visualisation. In this case, the discussion is characterised by equivocation and
ultimately defection by two players, some dissonance in alignment and negotiation, and
subtly contrasting discourse styles.
Let us start with the speech of the two defectors. In utterance 5, Gold1F structurally
aligns with Gold2M’s proposed collective strategy early on, resonating with his syntactic
construction (Du Bois & Giora 2014) and upgrading the collective benefit of altruism from
“you” to an inclusive “us”. This draws attention to her trustworthiness without an explicit
promise, and she does not participate further in the discussion. As Goffman (1956, p. 41)
notes, “communication techniques such as innuendo, strategic ambiguity, and crucial
omissions allow the misinformer to profit from lies, without, technically, telling any.” Gold1F
uses vagueness to avoid lying (much like Orange5F did), while other defectors in the study
were willing to manipulate their group members, as we will see with Gold5M. The upshot of
Gold1F’s comment is a perceived consensus on the part of Gold2M (cf. Bouas & Komorita
1996). In Round 2, when Gold1F’s defection was revealed in the Online Chat, he expressed
his anger through Multicultural London English (c.f. Cheshire et al. 2011): “GOLD1F is
moving mad!” before arguing that “Everyone agreed to share 40 to increase the pot - more
squeezes shared more money earned. someone is obviously content being selfish and earning
less, than sharing and earning more. (…) Que pasa!” Interestingly, at this later stage, Gold1F
responds with entirely explicit reasoning, and with none of her earlier structural alignment
now apparent: “Once you have calculated the maths, it allows you to make the right
contribution.”
The other defector, Gold5M, has proven himself in Round 1 by contributing 30
squeezes to the group. Like Gold1F, he triggers an “activation of affinities across utterances”
(Du Bois & Giora 2014, p. 356) in utterances 6, 9, and 12 to emphasize the precise financial
30

reward of cooperating. His strong epistemic forms (“exactly”, imperatives, direct
questioning) support an authoritative position within the group, as we saw in both previous
teams; in this case it also avoid arousing suspicion. In Round 2, Gold5M’s response to the
punishments from his disgruntled group shows the explicit insincerity of his utterances in
Round 1: “It benefits me when I convince everyone to contribute as much as possible while I
contribute none. This allows me to get the most. Sorry guys.”
In addition to these sincerity and non-participation issues, a further source of
dissonance in the exchange is an unsettled mix of registers and frames. Gold2M’s use of the
slang term “deffo” (‘definitely’) and an omitted pronoun and article conveys informality, as
does his ally Gold5M’s “yea” later. By contrast, Gold4M enters the exchange with a more
formal, business voice. The term “profitable”, the only use of the word in all 18 discussions,
and whose root (“profit”) features in the top 100 keywords of Nelson’s (2000) Business
English Corpus (BEC). We see an influence of business speak in later comments by Gold2M
and Gold5M, e.g., “maximize”, “reward”, “contribution”, “approve”, “best possible solution”
(once again, “solution” was found nowhere else in the discussions but featured in Nelson’s
(2000) list of positive keywords in the BEC. Gold4M’s comments are also marked by
weakeners “quite” and “we can” and a lack of affect display, all of which introduce
ambiguous personal commitment. Gold4M may benefit from this discursive distancing when
he later makes a counteroffer to Gold5M’s proposal.
Gold2M’s enthusiastic endorsement of Gold5M forms an alliance; unfortunately for
Gold2M, defectors can interpret pledges as a sign that the co-operator is easy target (Camera
et al. 2013). Gold5M conceals this well with his high degree of alignment in comment 10.
The final segment involves a negotiation between Gold2M/Gold5M and Gold4M, who is
eventually convinced to cooperate as indicated by his agreement the following afternoon.
31

Once the results are revealed in Round 2, Gold2M’s sense of betrayal is evident. He
not only scorns Gold1F, as we saw earlier, but also Gold5M: “GOLD5M is moving loose!”
His strong negative alignment is paralleled by a structural shift away from his brief use of the
business register to MLE. Gold3F’s response confirms the sense that group members
produced and perceived the discussion as formulaic of a contractual business meeting: ‘Well
done to those who kept to the agreement. Clearly some group members cannot help being out
for themselves!!’ Gold4M focused solely on the moral contract: “Greed can be seen from 2
guys here.” We end with this example to acknowledge that talk can be cheap. All the same,
Gold1F avoided explicit deception, which suggests a strong constraining role of reputation.
Combined with the binding promises of the remaining three players and their intense negative
reaction to Gold3M’s broken promise, talk is widely used even in this group as a sincere
signal of cooperative intentions. Finally, it is important to bear in mind that the groups
showed a fair level of correspondence between talk and behaviour overall (Table 2).
5. OTHER TEAMS
How representative are these case studies of the teams as a whole? Teams with low consensus
scores in Table 2 were negatively impacted by a combination of issues, starting with weak
leadership, which we have seen affected Orange team. Among the two most unsuccessful
groups, both of which contributed fewer squeezes at Round 2, discussions were dominated by
a single participant who then brought personal and group competency problems to the fore. In
Turquoise team, only one participant increased her contribution in Round 2, a non-native
female who participated heavily at the beginning of the discussion. Her leading role in
conversation and rather formal and stilted language resembled that of Orange3F. She
disaligned from the group with doubts over their altruism (“Also, at the very beginning I was
32

a bit skeptical on how much to share as I wanted to see how the rest of the group would
behave”), and this mistrust is cited later by a defector in Round 2 (“Oh wow…well done
Turquoise2F! I didn’t give more squeezes because I thought everyone was going to pull a 0
for the group pot in this round.”) Similarly, in the Scarlet team, 12 out of 18 comments were
from a female participant who was simultaneously figuring out the task and trying to guide
the group strategy. The leadership of these two participants contrasts dramatically with
Green1M’s adept and adaptive leadership style.
Other unsuccessful groups encountered related problems of distrust, low expectations
of cooperation (e.g., expressing surprise when presented with evidence of cooperation),
unfamiliarity with the format of chatroom discussion (e.g., awkwardly repeating previous
posts), and misaligned comments. These factors negatively impacted the consensus process in
two ways: the quantity and quality of participants’ promises to cooperate. This is supported
by the fact that in the bottom four groups, there were not only far fewer promises, but these
were also more vague in terms of commitment (e.g., a promise to contribute more rather than
all 40). A non-committal promise or no promise at all is a safe option when mistrust is
present, leadership is shaky, or understanding is low, and so these triggers of disalignment led
to an inability to reach a consensus to cooperate.
By contrast, the top four groups had a far higher number of promises, and more that
specified an exact contribution of 40 squeezes. Did this strong consensus arise in similar
ways to our case study of Green team above? The most successful group, Mauve, had an
initial misunderstanding that was resolved, after which all participants were highly aligned,
making five promises to contribute all 40 squeezes in Round 2. More importantly, several
used a gameshow register just like Green team, with comments such as “TEAM MAUVE!!!”
and “This reminds me of the show Golden Balls” (to which someone replied: “I’m more of a
Crystal Maze man myself.”) Cooperating teams showed further similarities to those discussed
33

here, including socialist discourse, resolution of misunderstanding, and playfully stern
exhortations to cooperate that received widespread uptake.
6. SUMMARY
Our analysis shows that communication does not have a fixed effect on cooperation.
Numerous elements are at play during spoken cooperation, and the relative success an actor
has in imposing a given strategy depends on uptake within the group. For example, one
successful group may fall in line with a dominant but persuasive leader, while another may
converge upon an egalitarian norm for stance convergence. A mixed methods approach
examining these qualitative underpinnings of a quantitative outcome helps to harness the
richness of natural language while operationalising it as a measurable factor in economic
behaviour.
Despite the non-deterministic nature of cooperation through talk, our study indicates
that certain registers are regularly associated with cooperative behaviour: gameshow, familial,
and socialist styles of interaction. These voices were used to align with or disalign from
others in three crucial ways: (1) foster group identity through encouragement and clarification
of misunderstandings; (2) instil a moral norm of cooperation by expressing high expectations
or delivering punishment; and/or (3) ensure players publicly commit to cooperate via strong
leadership. Even beyond our case studies, successful groups displayed these three strategies
to varying degrees (Yellow, Rose, Mauve), supporting details of past proposals (moral
norms, Bicchieri & Lev-On 2007; public commitment, Bouas & Komorita 1996, and group
identity, Orbell et al. 1988). The cases here show how many of these isolated factors work
together in natural spoken language.
By contrast, political speech and business talk were registers that featured in less
cooperative groups. The individualistic agenda of figures in the world of politics and business
34

may not only evoke associations of mistrust, sales pitches, and heavy-handed leadership, they
may also reduce obligation due to the high social distance of such voices. These effects were
observable in the three case studies as well as in other unsuccessful groups, along with low
expectations (Black, Silver) and unresolved confusion (Scarlet).
7. LIMITATIONS AND FUTURE WORK
A case study approach undoubtedly faces problems of generalisability and subjectivity. We
avoid claims of direct causation here, given the nature of the data, but we offer the
statistically significant consensus effect in the wider study (Table 2) as one way to link the
moment-to-moment negotiation of talk to behavioural outcomes. Understanding how talk
builds consensus may be challenging in methodological terms but must form part of the
scientific study of cooperative behaviour. Open-ended data collection provides us with
“inevitably richer” information than closed-ended quantitative methods (Starr 2014, p. 240)
and the two complement each other. Current quantitative methods in social dilemma research
include content analyses (Goren & Bornstein 2000; Kagel 2018; McClung et al. 2017) or
comparisons of specific types of communication, from repeated vs. one-off (Oprea et al.
2014; Ostrom et al. 1992) to binding vs. non-binding pledges (Chen & Komorita 1994).
These studies are valuable in identifying aggregate effects, but they do not attend to the
nuances of communication. We have proposed both quantitative (correlation of aggregate
consensus score with subsequent behaviour; degrees of inter-personal alignment in real-time)
and qualitative (choice of register and imposition of behavioural norms via frames) tools to
better understand the relationship between communication and behaviour. Further research
would benefit from alignment visualisations of unrestricted communication across a greater
35

number of groups to gauge whether there are identifiable patterns of stancetaking, as well as
additional qualitative analyses to uncover similar or different registers.
The study would also benefit from a brief comprehension task to ensure that all
participants understood the game prior to Round 1. The experimenter asked each participant
if the instructions were clear and if there were any questions. On occasion, they had doubts
which were addressed, but some participants may have been reluctant to express their
confusion. These misunderstandings were sometimes resolved in the group, for example,
Orange1F initially struggled to calculate what level of cooperation would be the most
beneficial for the group before Orange3F’s ‘speech’ highlighted the group benefit of full
cooperation. For other groups, such as the Scarlet team, it was precisely this
misunderstanding which dominated the discussion and potentially impacted the group’s
ability to reach a consensus to cooperate. However, the Online Chat in Round 1 and 2 for
most groups, including the 3 discussed here, indicate general understanding of the risk
involved in cooperating and the conflict between self vs. others’ interests (e.g., Green4F:
“Wow great job everyone, no one betrayed the group”; Orange3F: “…given this is
anonymous and we don’t know what everyone else to do, it’s hard to decide”).
We cannot rule out the simple factor of differences in individual personality driving
the above dynamics. There has been increased interest in explaining behavioural
heterogeneity through the HEXACO personality framework (Ashton & Lee 2007),
comprising Honesty-Humility, Emotionality, eXtraversion, Agreeableness, Conscientiousness
and Openness. Of particular relevance to social dilemmas is the Honesty-Humility (HH) scale
which includes greed avoidance, sincerity, fairness and modesty (Lee et al. 2008; Lee et al.
2010; Lee et al. 2013). Research suggests that fairness and greed avoidance are strongly
related to prosocial behaviour (Hilbig et al. 2014). Future work exploring verbal interaction
36

among people with similar levels of HH would allow us to tease apart the role of talk and
individual differences in cooperation.
It remains to be seen what communicative patterns would emerge in experiments with
shorter or longer time periods between rounds. Every participant was shown the discussion
before completing Round 2 to ensure equal exposure and recent emotional response to the
discussion. However, it is possible that difference in time between the Online Chat in Round
1 and Effort Task in Round 2 affected the intensity of emotions felt after the discussion, with
consequences for Round 2 behaviour. For example, a ‘miffed’ participant in the Amber team
admitted to the experimenter that she was initially very upset by her group’s behaviour, and
she may well have acted differently if she had completed Round 2 sooner.
Another fruitful line of enquiry would build on economic research that has attested to
the impact of medium on cooperation (Bicchieri & Lev-On 2007). The presence of non-
verbal cues in video-call settings, for example, would no doubt add nuance to expressions of
register and alignment between speakers, but would this help or hinder cooperation? This
question becomes increasingly relevant in the digital age where novel communication
technology is expanding the contexts in which social dilemmas are resolved.
We urge researches to explore these lines of enquiry given that crucial real-world
discussions which require cooperation are shifting online due to the exponential advancement
of technology. An important application of this research is harnessing Natural Language
Processing (NLP) to model cooperative behaviour. Ozkes and Hanaki (2020) adopt a machine
learning perspective to examine free-form communication in the context of cooperation. In
line with our incremental approach to decision-making, they conclude that subjects “reach
full cooperation by agreeing on gradual moves towards it” (p. 1). Training algorithms to
detect and predict cooperation based on a quantitative and qualitative analysis of
37

communication as outlined above would have far-reaching applications in the realm of
economics and beyond (cf. Kiesling et al. 2017).
Social dilemmas affect society at all levels, from international agreements to intimate
relationships. By exploring the registers available to individual actors and the alignments
between actors, we hope to enrich traditional studies on public goods games and generate
interdisciplinary dialogue between sociolinguistics and economics.
References
Agha, A., 2005. Voice, Footing & Enregisterment. Journal of Linguistic Anthropology 15(1),
38-59. https://doi.org/10.1525/jlin.2005.15.1.38.
Agha, A., 2015. Enregisterment and communication in social history. In: Agha, A., Frog.
(Eds.). Registers of Communication. Finnish Literature Society, 27-53.
https://doi.org/10.2307/j.ctvggx2qk.
Archakis, A., Tsakona, V., 2010. The wolf wakes up inside them, grows warewolf hair and
reveals all their bullying: The representation of parliamentary discourse in Greek
newspapers. Journal of Pragmatics 42, 912-923.
https://doi.org/10.1016/j.pragma.2009.08.015.
Ashton, M., Kibeom, L., 2007. Empirical, theoretical and practical advantages of the
HEXACO model of personality structure. Personality and Social Psychology Review.
11(2), 150-166. 10.1177/1088868306294907.
38

Author1., Author2., Author3. In preparation. Do as I say not as I do: The role of social
comparison in effort exertion.
Bakhtin, M., 1981. The Dialogic Imagination. Four Essays. Edited by Michael Holquist.
Translated by Caryl Emerson and Michael Holquist. Austin: University of Texas Press.
Balliet, D., 2010. Communication and cooperation in social dilemmas: A meta-analytic
review. Journal of Conflict Resolution 54(1), 39-57.
https://doi.org/10.1177/0022002709352443.
Bates, D., Maechler, M., Bolker, B., Walker, S., 2015. Fitting linear mixed- models using
lme4. Journal of Statistical Software 67(1), 1-48. 10.18637/jss.v067.i01.
Biber, D., Finegan, E., 1994. Sociolinguistic perspectives on register. Oxford: Oxford
University Press.
Bicchieri, C., 2002. Covenants without swords: Group identity, norms, and communication in
social dilemmas. Rationality and Society 14, 192-228.
https://doi.org/10.1177/1043463102014002003.
Bicchieri, C., Lev-On, A., 2007. Computer-mediated communication and cooperation in
social dilemmas: An experimental analysis. Politics, Philosophy & Economics 6(2),
139-168. https://doi.org/10.1177/1470594X07077267.
39

Bouas, K., Komorita, S., 1996. Group discussion and cooperation in social dilemmas.
Personality and Social Psychology Bulletin 22(11), 1144-1150.
https://doi.org/10.1177/01461672962211005.
Brown, P., Levinson, S., 1987. Politeness: Some universals in language usage. Cambridge,
United Kingdom: Cambridge University Press.
Burnett, H., 2019. Signalling games, sociolinguistic variation and the construction of style.
Linguistics and Philosophy 42, 419-450. https://doi.org/10.1007/s10988-018-9254-y.
Camera, G., Casari, M., Bigoni, M., 2013. Binding promises and cooperation among
strangers. Economic Letters 118, 459-461.
https://doi.org/10.1016/j.econlet.2012.12.009.
Charness, G., Dufwenberg, M., 2006. Promises and partnership. Econometrica 74(6), 1579-
1601. https://doi.org/10.1111/j.1468-0262.2006.00719.x.
Chen, X., Komorita, S., 1994. The effects of communication and commitment in a public
goods social dilemma. Organizational Behaviour and Human Decision Processes 60,
367-386. https://doi.org/10.1006/obhd.1994.1090.
Cheshire, J., Kerswill, P., Fox, S., Torgeresen, E., 2011. Contact, the feature pool and the
speech community: The emergence of Multicultural London English. Journal of
Sociolinguistics 15(2), 151-196. https://doi.org/10.1111/j.1467-9841.2011.00478.x.
40

Coreen, F., Sandler, S., 2014. Polyphony, ventriloquism and constitution: In dialogue with
Bakhtin. Communication Theory 24(3), 225-244. https://doi.org/10.1111/comt.12041.
Dawes, R., McTavish, J., Shaklee, H., 1977. Behaviour, communication and assumptions
about other peoples’ behaviour in a commons dilemma situation. Journal of Personality
and Social Psychology 35, 1-11. https://doi.org/10.1037/0022-3514.35.1.1.
Du Bois, J., 2007. The stance triangle. In Stancetaking in discourse, ed. Robert Englebretson,
139–182, Amsterdam: John Benjamins. 10.1075/pbns.164.07du.
Du Bois, J., 2011. Co-opting intersubjectivity: Dialogic rhetoric of the self. In: Meyer, C.,
Girke, F (Eds.). The rhetorical emergence of culture. Oxford: Berghahn, 52-83.
10.5040/9781474215442.ch-013.
Du Bois, J., Giora, R., 2014. From cognitive-functional linguistics to dialogic syntax.
Cognitive Linguistics 25(3), 351-357. https://doi.org/10.1515/cog-2014-0023.
Du Bois, J., Kärkkäinen, E., 2012. Taking a stance on emotion: Affect, sequence, and
intersubjectivity in dialogic interaction. Text & Talk 32 (4), 433-451.
https://doi.org/10.1515/text-2012-0021.
Duranti, A., 2006. Narrating the political self in a campaign for U.S. Congress. Language in
Society 35, 467-497.
41

Dunning, D., Anderson, J., Schlosser, T., Echlebracht, D., Fetchenhauer, D., 2014. Trust at
zero acquaintance: More a matter of respect that expectation of reward. Journal of
Personality and Social Psychology 107(1), 122-141. https://doi.org/10.1037/a0036673.
Engelmann, J., Fehr, E., 2017. The neurobiology of trust: The important role of emotions. In
van Lange, P., Rockenbach, B., Yamagashi, T (Eds.), Social Dilemmas: New
Perspectives on Reward and Punishment. New York: Oxford University Press, 33-56.
Fezter, A., 2014. I think, I mean, I believe in political discourse. Collocates, functions and
distribution. Functions of Language 21(1), 67-94. https://doi.org/10.1075/fol.21.1.05fet.
Fetzer, A., Bull, P., 2012. Doing leadership in political speech: Semantic processes and
pragmatic inferences. Discourse and Society 23(2), 127-144
https://doi.org/10.1177/0957926511431510.
Franke, M,. 2009. Signal to act: Game theory in pragmatics. Institute for Logic, Language
and Computation.
Gachter, S., Fehr, E. 1999. Collective action as a social exchange. Journal of Economic
Behaviour & Organisation 39(4), 341-369. https://doi.org/10.1016/S0167-
2681(99)00045-1
Goffman, E., 1956. The presentation of self in everyday life. Edinburgh: University of
Edinburgh.
42

Goffman, E., 1961. Two studies in the sociology of interaction. Indianapolis: Bobbs-Merrill.
Goffman, E., 1974. Frame analysis: An essay on the organisation of experience. Harvard
University Press.
Goodman, N., Frank, M., 2016. Pragmatic language interpretation as a probabilistic
inference. Trends in Cognitive Sciences 20(11), 818-829.
DOI:https://doi.org/10.1016/j.tics.2016.08.005.
Goren, H., Bornstein, B., 2000. The effects of intragroup communication on intergroup
cooperation in the repeated prisoner’s dilemma. Journal of Conflict Resolution 44(5).
700- 719. https://doi.org/10.1177/0022002700044005007.
Halvorsen, K., 2018. Questions as interaction resource in team decision making. International
Journal of Business and Communication 55(1), 69-73.
https://doi.org/10.1177/2329488415589102.
Hilbig, B., Glockner, A., Zettler, I., 2014. Personality and prosocial behaviour: Linking basic
traits and social value orientations. Journal of Personality and Social Psychology
107(3), 529-539. https://doi.org/10.1037/a0036074.
Houghton, K., Upadhyay, S., Kin, C., 2018. Punctuation in text messages may convey
abruptness. Period. Computers in Human Behaviour 80, 112-121.
https://doi.org/10.1016/j.chb.2017.10.044.
43

Ip, A., 2002. The impact of emoticons can affect interpretation in instant messaging. Carnegie
Mellon University. Retrieved from http://amysmile.com/pastprj/emoticon_paper.pdf.
Accessed on 2nd December 2020.
Jacknick, C., Avni, S., 2016. Shalom, bitches: Epistemic stance and identity work in an
anonymous online forum. Discourse, Context & Media 15, 54-64.
https://doi.org/10.1016/j.dcm.2016.11.002.
Jaffe, A., 2009. The sociolinguistics of stance. In Alexandra Jaffe (Ed.). Stance:
Sociolinguistics Perspectives. New York: Oxford University Press, 3-28.
Jäger, G., 2008. Applications of game theory in linguistics. Language and Linguistics
Compass 2(3), 406-421. 10.1111/j.1749-818X.2008.00053.x.
Joseph John. 2006. Language and Politics. Edinburgh: Edinburgh University Press.
Kagel, J., 2018. Cooperation through communication: Teams and individuals in finitely
repeated Prisoner’s dilemma games. Journal of Economic Behaviour & Organization
146, 55-64. https://doi.org/10.1016/j.jebo.2017.12.009.
Kärkkäinen, E., 2006. Stancetaking in conversation: From subjectivity to intersubjectivity.
Text & Talk 26(6), 699-731. https://doi.org/10.1515/TEXT.2006.029.
44

Kiesling, S., Pavalanathan, U., Fitzpatrick, J., Han, X., Eisenstein, J., 2017. Interactional
stancetaking in online forums. Computational Linguistics 44(4), 683-718
https://doi.org/10.1162/coli_a_00334.
Kerr, N., Kaufman-Gilliland, C., 1994. Communication, commitment and cooperation in
social dilemma. Journal of Personality and Social Psychology 66(3), 513- 529.
https://doi.org/10.1037/0022-3514.66.3.513.
Koester, Almut. 2010. Workplace Discourse. London: Continuum.
Lee, K., Ashton, M., Morrison, D., Cordery, J., Dunlop, P., 2008. Predicting integrity with the
HEXACO personality model: Use of self- and observer reports. Journal of
Occupational and Organizational Psychology, 81, 147–167.
https://doi.org/10.1348/096317907X195175.
Lee, K., Ashton, M., Ogunfowora, B., Bourdage, J., Shin, K-H., 2010. The personality bases
of socio-political attitudes: The role of Honesty-Humility and Openness to experience.
Journal of Research in Personality, 44, 115–119.
https://doi.org/10.1016/j.jrp.2009.08.007.
Lee, K., Ashton, M., Wiltshire, J., Bourdage, J., Visser, B., Gallucci, A., 2013. Sex, power,
and money: Prediction from the Dark Triad and Honesty- Humility. European Journal
of Personality, 27, 169–184. https://doi.org/10.1002/per.1860.
45

Lenger, A., 2019. The rejection of qualitative research methods in economics. Journal of
Economic Issues 53(4), 946-965. https://doi.org/10.1080/00213624.2019.1657748.
Levinson, S., 1979. Activity types and language. Linguistics 17, 365-399.
https://doi.org/10.1515/ling.1979.17.5-6.365.
Majeski, S., Fricks, S., 1995. Conflict and cooperation in international relations. Journal of
Conflict Resolution 39(4), 622-645. https://doi.org/10.1177/0022002795039004002
McClung, J., Placi, S., Bangerter, A., Clement, F., Bshary, R., 2017. The language of
cooperation: Shared intentionality drives variation in helping as a function of group
membership. Proceedings of the Royal Society B. Biological Sciences 284(1863).
https://doi.org/10.1098/rspb.2017.1682.
Milkovich M., Sitarica, A., 2017 Epistemic modality in political discourse. Research Result,
Social Studies and Humanities. 3(1), 75-79 DOI: 10.18413/2408-932X-2017-3-1-75-
79.
Muehlbacher, S., Kirchler, E., 2009. Origin of endowment in public good games: The impact
of effort on contributions. Journal of Neuroscience, Psychology and Economics 2(1),
59-67.
Myers-Scotton, C., Bolonyai, A., 2001. Calculating speakers: Codeswitching in a rational
choice model. Language in Society 30(1), 1-28. https://doi.org/10.1037/a0015458.
46

Nelson, M., 2000. Corpus-based study of the lexis of Business English and Business English
Teaching Material. Unpublished Thesis, University of Manchester, Manchester.
Retrieved 17th July 2020 from http://users.utu.fi/micnel/thesis.html.
Ochs E., Taylor, C., 1995. The “Father knows best” dynamic in dinnertime narratives. In:
Hall, K., Bucholtz, M. (Eds.). Gender articulated: Language and the socially
constructed Self. London: Routledge, 97-120.
Oprea, R., Charness, G., Friedman, D., 2014. Continuous time and communication in a public
goods experiment. Journal of Economic Behaviour and Organisation 108, 212- 223.
https://doi.org/10.1016/j.jebo.2014.09.012.
Orbell, J., Van de Kragt, A., Dawes, R., 1988. Explaining discussion-induced cooperation.
Journal of Personality and Social Psychology 54(5), 811-819.
https://doi.org/10.1037/0022-3514.54.5.811.
Ostrom, E., Walker, J., Gardner, R., 1992. Covenants with and without a sword: Self-
governance is possible. The American Political Science Review 86(2), 404-417.
doi:10.2307/1964229.
Ozkes, A., Hanaki, N., 2020. Strategic environment effect and communication. Available at
SSRN: https://ssrn.com/abstract=3484927.
Perry, I., 2004. Prophets of the hood: Politics and poetics in Hip Hop. Durham & London:
Duke University Press.
47

Roe, K., Messing, V., Vandebosch, H., Van den Bulck, J., 1996. Television game show
viewers: A cultivated audience? Communications 21(1), 49-64.
https://doi.org/10.1515/comm.1996.21.1.49.
Sally, D., 2000. A general theory of sympathy, mind-reading, and social interaction, with an
application to the prisoners’ dilemma. Social Science Information 39, 567–634.
https://doi.org/10.1177/053901800039004003.
Sally, D., 2005. Can I say “bobobo” and mean “There’s no such thing as cheap talk”? Journal
of Economic Behaviour and Organization 57, 245-266.
https://doi.org/10.1016/j.jebo.2004.02.004.
Searle, J., 1969. Speech acts: An essay in the philosophy of language. Cambridge: Cambridge
University Press.
Shank D., Kashima, Y., Peters, K., Li, Y., Robins, G., Kirley, M., 2019. Norm talk and human
cooperation: Can we talk ourselves into cooperation? Journal of Personality and Social
Psychology 117(1), 1-25. DOI: 10.1037/pspi0000163.
Starr, M., 2014. Qualitative and mixed methods research in economics: Surprising growth,
promising future. Journal of Economic Surveys 28(2), 238-264.
https://doi.org/10.1111/joes.12004.
48

Tajfel, H., Turner, J., 1979. The social identity theory of intergroup behavior. In: Williams, A.,
Worchel, S. (Eds.). The Social Psychology of Intergroup Relations. Chicago: Nelson‐
Hall. 7–24.
Twenge, J., Campbell, K., Gentile, B., 2012. Increases in individualistic words and phrases in
American books, 1960-2008. PLoSONE 7(7). doi: 10.1371/journal.pone.0040181.
van der Valk, I., 2003. Right-Wing parliamentary discourse on immigration in France.
Discourse & Society 14 (3), 309-348. https://doi.org/10.1177/09579265030143004
van Leeuwen, T., 2007. Legitimation in discourse and communication. Discourse and
Communication 1(1), 91-112. https://doi.org/10.1177/1750481307071986.
Wilson, R., Sell, J., 1997. “Liar liar...”: Cheap talk and reputation in repeated public goods
settings. Journal of Conflict Resolution 41(5), 695-717.
https://doi.org/10.1177/0022002797041005005.
Zheng, J., Bos, N., Olson, J., Olson, G., 2002. Trust without touch: Jump-start trust with
social chat. Proceedings of Human Factors in Computing Systems 2001 Extended
Abstracts, 291- 292. New York: ACM. https://doi.org/10.1145/634067.634241.
49

50