Temple University Persuasive Campaign Essay

Read the attached matierials about heuristics and biases, then think about the anti-vaping paper topic and do the following: (a) recommend one design feature of your communication effort and (b) explain which of the heuristics or biases informs your recommendation about campaign design and how.  Be sure to choose one (and only one) of the heuristics or biases.

Available online at www.sciencedirect.com
Journal of Pragmatics 59 (2013) 190–199
www.elsevier.com/locate/pragma
Constraining context selection: On the pragmatic inevitability
of manipulation
Didier Maillat *
University of Fribourg, Switzerland
Received 11 May 2012; received in revised form 5 June 2013; accepted 22 July 2013
Abstract
In this paper, I propose a new take on manipulation that focuses on the hearer’s cognitive processing of manipulative utterances,
thereby avoiding some of the descriptive and explanatory limitations of other approaches to manipulation (see de Saussure and Schulz,
2005 for discussion). I will follow Sperber and Wilson (1995) and posit that interpretation is a context building process through which
contextual assumptions are incrementally added to an interpretative context subset.
The main argument in this paper presents an analytic model that captures the specificity of manipulative strategies as attempts to
exploit the inherent limitations of these context selection processes. Specifically, manipulation will be shown to be best accounted for as a
form of communication that tries to interfere with the order of accessibility of contextual assumptions to ensure that the hearer’s
interpretation of a manipulative utterance only accesses a sub-optimal set of contextual assumption, crucially preventing her from
accessing a dissonant, although optimal, set (see Maillat and Oswald, 2009).
The pragmatics of such interpretative biases, it will be argued, is best accounted for within a relevance-theoretic framework, as the
model posits that the cognitive mechanisms which govern interpretation are error-prone as a result of our ‘cognitive optimism’ (Sperber
et al., 1995, 2010).
© 2013 Elsevier B.V. All rights reserved.
Keywords: Argumentation; Manipulation; Cognitive illusion; Relevance theory; Propaganda; Fallacy
1. Introduction
As pragmatics studies the construction of meaning in use, its scope inevitably encompasses manipulative discourse and
as a result the paths of manipulation and pragmatic analysis have crossed on several occasions.1 In fact, in most instances
manipulation has been regarded as a form of deficient communication in some respect. Authors point out the ill-formed
nature of manipulative utterances on various — not necessarily compatible — levels. The purpose of this paper is to propose a
pragmatic model of manipulative discourse that couches in pragmatic theory the specificity of the interpretative processes
that govern manipulation. In this respect, this paper attends to the analytic void noted by Cummings (2005) when she
underlined the urgent need for a principled pragmatic theory of argumentation. Thus, by looking at some reputedly ill-formed
arguments — manipulative moves — a first glimpse of what such a principled theory would consist of will be presented.
In doing so, I will place the main focus of this study on the comprehension component of communication and will
suggest that an appropriate definition of manipulation should focus on the hearer’s side of the communicative exchange.
* Correspondence to: Department of English, University of Fribourg, Europe 20, CH-1700 Fribourg, Switzerland. Tel.: +41 26 300 79 07; fax: +41
26 300 97 87.
E-mail address: didier.maillat@unifr.ch.
1
I would like to thank two anonymous reviewers for their feedback and suggestions, as well as Steve Oswald for the stimulating discussions we
invariably have on these questions.
0378-2166/$ — see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.pragma.2013.07.009
D. Maillat / Journal of Pragmatics 59 (2013) 190–199
191
Manipulation will thus be viewed primarily as something that affects a hearer. In the course of establishing this model the
analytic weight will also be shifted from an attempt to define manipulation negatively as communication minus some
component to an attempt at defining manipulation as a predictable — inevitable — development of the normal pragmatic
processes that govern everyday communication. It will be argued that Relevance Theory is better equipped to scrutinise
the sort of mechanisms that ensure the success of manipulative discourse. In the last part of this paper, I will analyse a
well-known manipulative move, the argument ad populum, to claim that while undoubtedly problematic, this type of
argument is in effect perfectly valid from an interpretative and cognitive point of view.
2. Defining manipulation
Manipulation is often presented as an ill-formed type of communication. Many of the existing attempts to characterise
manipulation have tried to define manipulation as deficient in some respect (see Maillat and Oswald, 2009 for a more
detailed discussion). The actual nature of this defect varies. Rigotti (2005) for instance looked into the truth-conditional
limitations of manipulative arguments. Lies would prototypically illustrate such manipulative uses. However, such a
definition is too limiting on at least two counts: first truth-conditional inconsistencies do not capture other forms of
manipulative arguments, for instance the use of fear to strengthen one’s argument (as in the case of a threat I wouldn’t do
that if I were you or that of an argument ad baculum). Second, even in cases where arguably truth-conditions are blatantly
violated, as for example with an ad populum argument of the type (1) Everybody thinks that Romney will be the Republican
candidate, it seems that the gist of the manipulative intention lies elsewhere then in its attempt at misrepresenting the truth
(i.e. not everybody thinks that). In our example, the strength of the manipulative use seems to be retained if we limit the
referential scope of the universal pronoun: (2) Everybody I spoke to about this thinks that Romney will be the Republican
candidate. (2) is still a manipulative use that presents the number of speakers who express a thought as an argument in
favour of that thought’s inherent validity, ignoring the fact that everybody, respectively everybody-I-spoke-to-about-this,
could be wrong.
Other accounts, such as van Dijk’s (2006) are derived from the tradition in critical discourse analysis, which focuses on
power relations as they are negotiated and imposed through discourse. In this line of research, manipulative discourse is
primarily conceived of as involving an effort on the speaker’s part to guarantee her own dominant social position by
pretending that an argument is in the hearer’s interest, when it really serves the interests of the speaker (see also the
discussion in de Saussure, 2005). Such a definition also appears to be too restrictive in that it excludes many
communicative situations in which a speaker wishes to manipulate the hearer to think X for his own good.2
Closer to the pragmatic turf, scholars have tried to elucidate the specificity of manipulative discourse using typically
pragmatic tools. Parret (1978) for instance evaluated the possibility of a speech-act-theoretic approach, where the illformedness of manipulative moves is discussed in terms of a violation of felicity conditions. However, whether you wish to
couch manipulation in terms of a perlocutionary effect or of an illocutionary force, the very idea of a category of
manipulative speech acts (alongside warnings, promises, etc.) is self-defeating as a it would be a type of speech act that
would cease to function the very moment it is identified (see the discussion in Parret, 1978).
Finally, in an attempt based on a similar set of assumptions to the one proposed in the following section, some scholars
emphasised the discrepancy between the manipulative speaker’s communicative goal and the hearer’s retrieval of her
intention (Attardo, 1997; Rigotti, 2005). Such an account is bound to encounter some difficulties however, as it seems to
define manipulation as something that a speaker does when in fact it can be shown that manipulation can be achieved by
a set of utterances independently of the speaker who produces them.
For example, we can think of a situation where a set of manipulative arguments achieve the same manipulative
effect on a hearer independently of who relayed them to the hearer, as in the case of racist fallacious arguments that
rely on anecdotal evidence to justify racist laws which validate social inequalities based on racial ancestry. It might be
argued that the initiator of such propagandist communication is genuinely fooling her audience about her hidden
intention. Such an intention could be paraphrased in the case in point, as an attempt at creating a sense of bonding
around a feeling of self-professed superiority and a form of animosity towards social minorities. But such a set of
fallacious arguments can be repeated by speakers whose intentions would not match the hidden goal described here.
In this context, it would be difficult to analyse the manipulative use of fallacious arguments in terms of a misreading of
the speaker’s intention.
Back in 1933, the Nazi leaders knew that if a false statement was repeated often enough it could become the ‘truth’ in
the eyes of the German population. Obviously, they would never go as far as to admit it bluntly, but it is quite clear in their
writings. Thus, in a speech delivered as early as January 1928, Goebbels, who masterminded nazi propaganda, was
2
In this article, following RT’s practice, the speaker is conventionally assumed to be a woman and the hearer to be a man.
192
D. Maillat / Journal of Pragmatics 59 (2013) 190–199
saying that what matters in propaganda is not so much that some creed is generally true but that it be held true by the
people who will fight for their leader’s Weltanschauung.3 In fact, Hitler himself was even more transparent in his writings on
the relative advantages of using lies in propaganda. He insisted that useful lies not only could but should be put to the
service of the higher cause through propaganda. While discussing the failings of war propaganda in Germany during the
First World War, Hitler is keen to highlight that Allies’ propaganda was far superior for its systematic repetition of untruths.
Thus the British soldier was never allowed to feel that the information which he received at home was untrue.
Unfortunately the opposite was the case with the Germans, who finally wound up by rejecting everything from home
as pure swindle and humbug. [. . .] The worst of all was that our people did not understand the very first condition
which has to be fulfilled in every kind of propaganda; namely, a systematically one-sided attitude towards every
problem that has to be dealt with.[. . .] The aim of propaganda is not to try to pass judgment on conflicting rights,
giving each its due, but exclusively to emphasize the right which we are asserting. Propaganda must not investigate
the truth objectively and, in so far as it is favourable to the other side, present it according to the theoretical rules of
justice; [. . .] (translated by Murphy, 1939: chapter VI)
In this excerpt it is quite clear that truth is regarded as accessory in war propaganda and can even be detrimental to its
ultimate purpose. This apology of lies is underlined again when he points out that the Allies’ use of lies is to be emulated.
Further, the way in which they pilloried the German enemy as solely responsible for the war — which was a brutal and
absolute falsehood — and the way in which they proclaimed his guilt was excellently calculated to reach the masses,
realizing that these are always extremist in their feelings. And thus it was that this atrocious lie was positively
believed. The effectiveness of this kind of propaganda is well illustrated by the fact that after four-and-a-half years,
not only was the enemy still carrying on his propagandist work, but it was already undermining the stamina of our
people at home. (ibid.)
Finally, he insists on the importance of repetition to reach one’s propagandist aims. In one of the many allusions to the
paramount importance given in nazi propaganda to the repetition of the official creed, he thus writes that ‘‘only constant
repetition will finally succeed in imprinting an idea on the memory of the crowd’’ (ibid.).
It seems clear that while such a manipulative strategy (see the discussion of the ad populum fallacy below) would
reflect a hidden intention on the Nazi leaders’ part when they masterminded the propaganda around the blaze at the
Reichstag,4 such a hidden intention would be absent from the mind of most of the speakers who contributed to the rapid
spread of this manipulative piece of discourse.5 Arguably, though, it seems intuitively appropriate to claim that whoever
had been exposed to that manipulative strategy and had bought into it, that that person had been manipulated
(independently of the speaker’s intention). Such an example shows that manipulation can be passed on even if the
original, manipulative intention disappears in the process.
As should have become clear in this rapid survey, a recurrent problem with all of these accounts is that they focus on an
analysis of manipulation from the speaker’s perspective as most of them define it in terms of some hidden — and hence
unobservable — move or intention on her part. In the following I propose that a potentially richer analysis of manipulative
uses of language can be arrived at by reversing the viewpoint on these phenomena and by focusing on the hearer’s
interpretative processes.
3. The pragmatics of manipulation
A pragmatic approach to manipulative uses of discourse can be presented as an alternative to the limitations noted
above. More specifically, the present account purports to define manipulation as an effect on an addressee rather than as
misleading use of language on the speaker’s part. With such a shift in the perspective, I will claim that manipulative uses
become more accessible from an analytic point of view. They will also more readily lend themselves to observation,
contrary to what a phenomenon, like speaker-based manipulation, which is said to be covert.
Interestingly, such a shifted perspective in the pragmatic analysis leads to non trivial changes in the kind of definition
put forward for manipulative discourse. Perhaps the most important of these changes is linked to the communicative
nature of manipulation. On the hearer’s side, manipulative arguments are processed like any other regular utterance. In
3
‘Es kommt, wenn ich etwas als richtig erkannt habe, nicht darauf an, dass es in aller Theoretisiererei richtig ist, sondern vielmehr darauf, dass
ich Menschen finde, die mit mir für diese als richtig erkannte Weltanschauung zu kämpfen bereit sind.’ (Goebbels, 1938).
4
Communist activists were (erroneously) designated as the only culprits.
5
This survey is not intended as an exhaustive list. Other accounts have been put forward to account for manipulative uses of language, as
discussed in Maillat and Oswald (2009). The interested reader should turn to this reference for a more detailed discussion, as well as to the
volume edited by de Saussure and Schulz (2005).
D. Maillat / Journal of Pragmatics 59 (2013) 190–199
193
other words, the interpretative processes involved in the understanding of a manipulative utterance are exactly those
described under ‘normal’ circumstances. In fact, as will be argued below, it is precisely the interpretative unmarkedness of
manipulative utterances that ensures their success.6 This should not be taken to mean that the interpretation of
manipulative discourse cannot be distinguished from non-manipulative cases. The idea defended in this paper is to define
manipulation as an instance of sub-optimal output of the normal interpretation process, i.e. a case where a normal
interpretative procedure goes astray. The difference between this form of erroneous interpretative process and accidental
interpretative mistakes7 lies in the fact that this sub-optimal output is assumed to be the result of a prior constraint applied
onto the mechanisms that govern interpretation.
In order to describe this phenomenon, it needs to be considered within a pragmatic framework. There are several
reasons why Relevance Theory (Blakemore, 2002; Carston, 2002; Sperber and Wilson, 1995; Wilson and Sperber, 2012)
appears to be better suited at handling the sort of analysis that is required here. The first argument that speaks in favour of
the relevance-theoretic framework touches on its treatment of context. As Sperber and Wilson (1995) indicate, one of the
points of divergence with other pragmatic models rests in the way context is regarded as entering the interpretative
process. They write that ‘‘It is not that first context is determined, and then relevance is assessed. [. . .] [I]t is relevance
which is treated as given, and context which is treated as a variable’’ (1995:141–142). They explain further that context is
the result of the interpretative process.
Interestingly for an account of manipulative uses, by refocusing the interpretative process on context selection
procedures, they also predict that interpretation depends on these selection procedures and that if there were strategies to
control context selection procedures, the latter would amount to effectively controlling the hearer’s interpretation. In other
words, in a relevance-theoretic framework, manipulation could be analysed as an attempt to control the hearer’s context
selection process. This constitutes the analytic option explored in this paper.
Thus, relevance is not the goal of the interpretative process but rather the guiding principles that governs context
selection (1995:141–142): an optimally relevant interpretation minimises the cognitive efforts the hearer has to put into the
construction of the context set and maximises the cognitive effects generated via this context set. Crucially, as Sperber
and Wilson (1995) point out, the construction of the set of contextual assumptions is an incremental process, in which the
hearer accesses elements of his cognitive environment following an order of accessibility. That is to say that not all pieces
of information that are manifest to an individual at the time when he processes an utterance are equally accessible, some
assumptions are more manifest than others (1995:39–40). As the quest for optimal relevance only selects a small portion
of all the assumptions manifest to a hearer, it follows that given the very unlikely situation in which two addressees of the
same utterance would have two identical cognitive environments in terms of the assumptions that are manifest to them,
they might still differ with respect to the degree of manifestness (or accessibility) to which these assumptions are activated
in their cognitive environment. As a result, the same utterance U could lead to a different interpretation for these two
individuals simply because different assumptions will be selected first depending on their degree of manifestness.
Sperber and Wilson (1995) seem to capture the effect of an ordering relation in the cognitive environment using various
notions that seem to overlap at times. They write about degrees of manifestness, accessibility (1995:138), and also about
the strength of an assumption (1995:75ff, 143, 151). The fine-grained implementation of this ordered structure is not of
importance at this stage for the purpose of this discussion, however. What seems important to the understanding of
manipulative uses is the fact that
The organisation of the individual’s encyclopaedic memory, and the mental activity in which he is engaged, limit the
class of potential contexts from which an actual context can be chosen at any given time. [. . .] [N]ot all chunks of
encyclopaedic information are equally accessible at any given time (Sperber and Wilson, 1995:138).
Since the optimisation procedure that obeys the principle of relevance is not exhaustive,8 differences in the ordering
relation will lead to the selection of a different context set, and hence, to a different interpretation. Manipulation, it is argued, is
a discursive use that results in an effective re-ordering of contextual assumptions that affects their accessibility in the
cognitive environment of the hearer. In the above excerpt taken from Mein Kampf, Hitler explicitly describes this re-ordering
when he underlines ‘‘the very first condition which has to be fulfilled in every kind of propaganda; namely, a systematically
one-sided attitude towards every problem that has to be dealt with’’ (translated by Murphy, 1939: chapter VI).
The second reason that speaks in favour of a relevance theoretic framework is linked to the clear specification in the
model of a point at which pragmatic processes stop and yield an interpretation that is regarded as optimal. Crucially,
‘Interpretative unmarkedness’ refers to the fact that interpretation in cases of manipulative discourse does not depart from the norm.
Such a precision is needed as we do not want to treat as a case of manipulation an erroneous interpretation of (i) Max has this thing with bats
as meaning that Max likes baseball, when it is small flying mammals that Max cares about.
8
‘‘A phenomenon may make manifest a very large number of assumptions. However, this is not to say that the individual will actually construct
any, let alone all, of these assumptions’’ (Sperber and Wilson, 1995:151).
6
7
194
D. Maillat / Journal of Pragmatics 59 (2013) 190–199
Relevance Theory insists on the fact that relevance is calculated as an optimal ratio between cognitive efforts and effects,
but the notion of optimality used in this definition is not regarded as absolute. It is relative to a given cognitive environment,
as well as to a specific allocation of cognitive resources. Thus, the first principle of relevance is described as a tendency
towards optimal relevance (see discussion in Sperber and Wilson, 1995:262). But the authors hasten to add that
There may be many shortcomings, many cognitive sub-mechanisms that fail to deliver enough effect for the effort
they require, many occasions when the system’s resources are poorly allocated. (Sperber and Wilson, 1995:262)
In other words, the pragmatic system as presented in Relevance Theory can err at times. Not only can it err, but the theory
makes a certain number of predictions as to the type of errors that might lead the system astray and yield a sub-optimal
interpretation. The central claim of this paper argues that manipulative discourse exploits the inherent weaknesses of the
interpretative process to ensure that a sub-optimal interpretation is indeed arrived at, i.e. to ensure that one of the
predicted errors occurs. In this context, this approach looks at manipulative uses as a built-in — and hence, inevitable -consequence of the way our pragmatic system operates.
Specifically, the analysis proposed here suggests that a manipulator will achieve her goal by having a re-ordering action
on the cognitive environment of the hearer so as to guarantee that a given utterance U will be interpreted within an appropriate
subset of contextual assumptions, independently of the expected presence of contradictory assumptions in the cognitive
environment of the hearer. As a result, from an interpretative perspective, the main component of a manipulative argument is
a constraint applied on the selection process of contextual assumptions when interpreting a target utterance U. Manipulation
is therefore re-analysed as an instance of Context Selection Constraint (henceforth, CSC).
Manipulation as Context Selection Constraint — definition:
Manipulative communication is a twofold process by which a constraint that limits context selection is combined with
a target utterance U in order to force the interpretation of the latter within a limited set of contextual assumptions and
to effectively ensure that the interpretation is reached before a known, alternative (contradictory) subset of
assumptions is accessed.9
In this type of analysis, a manipulative use can adopt two different kinds of constraining patterns. According to Sperber
and Wilson (1995:140–141), the cognitive environment of a hearer is an ordered structure that obeys a partial inclusion
relation. In this ordered structure, the ‘‘assumptions left over in the memory of the deductive device, i.e. the interpretation
of the previous utterance, constitute an immediately given context’’. This initial level can be extended to a wider context set
that includes ‘‘the encyclopaedic entries (or possibly smaller chunks of encyclopaedic information, taken from these
entries) of concepts already present either in the context or in the assumption being processed’’. A third extension occurs
when ‘‘information about the immediately observable environment’’ is added. Most importantly for this approach,
[t]his formal relation has a psychological counterpart: order of inclusion corresponds to order of accessibility [. . .];
contexts which include only the initial context as a sub-part [. . .] are therefore the most accessible contexts; contexts
which include the initial context and a one-step extension as sub-parts [. . .] are therefore the next most accessible
contexts, and so on. (1995:142)
Going back to manipulation, the approach suggested here posits that manipulative uses ensure that a context Cˈ which
would be relevant to the interpretation of the target utterance U is not accessed, as it would yield a contradiction for U, and
would ultimately lead to the elimination of U from the cognitive environment of the hearer (CEH). Instead, manipulation
works on making a context C, in which U is strengthened and gives rise to contextual implications, much more accessible,
thereby effectively blocking access to Cˈ, since ‘‘[t]he less accessible a context, the greater the effort involved in accessing
it, and conversely’’ (1995:142). In other words, manipulation exploits the fact that relevance is more likely to be reached
within highly accessible contexts.
The reader will have noticed already that the difficulty with an analysis of manipulation that concentrates on the hearer
and does away with the speaker’s covert strategy mentioned in previous accounts resides in its ability to contrast
manipulation with normal interpretative phenomena; or, to set this difficulty within the framework of argumentation theory
(van Eemeren and Grootendorst, 2004), to contrast manipulation with persuasion.
In this account, persuasion is a discursive attempt at strengthening an utterance U by ensuring that U is consistent and,
ideally, strengthened in all relevant subsets of CEH.
:9 Cˈ , Cˈ  CEH and Cˈ≠C, such that U is relevant in Cˈ and U is inconsistent and eliminated in Cˈ
9
It is worth pointing out that this twofold process can take place over the course of two distinct utterances which may or may not be adjacent (as
in the case of affirming the consequent (see example below); or both steps can occur within one utterance as in the case of the standard ad
populum fallacy (Everybody says that U).
D. Maillat / Journal of Pragmatics 59 (2013) 190–199
195
That is to say that persuasion ensures that there is no contextual subset Cˈ in which U is both relevant and inconsistent
with other contextual assumptions. When the hearer is persuaded of U, the corresponding contextual assumption is
reinforced in all subsets of CEH in which U is relevant.
Whereas manipulation is a discursive attempt at guaranteeing that any known relevant subset Cˈ of CEH, which would
lead to the elimination of U, is not accessed or at least that it be less accessible than C.10
8Cˈ, Cˈ  CEH and Cˈ≠C, such that U is relevant in Cˈ and U is inconsistent and eliminated in Cˈ, accessibility (C)
> accessibility (Cˈ) in CEH
In other words, where persuasion excludes the existence of any contradicting contextual subset, manipulation, on the
other hand, makes sure that such a contradicting contextual subset, which is known to exist, is not accessed during the
interpretation of U.
The two definitions given above provide us with an interesting perspective on the contrast between the purported
hearer-based approach and earlier speaker-based accounts of manipulation. Thus, it should be clear from the above that
the proposed model does not try to dispense with the idea that a manipulative intention is at the origin of a manipulative
argument. However, the crucial step consists in re-analysing that intention in terms of the effect sought from the point of
view of the hearer’s interpretative processes. In such a framework, although an initial intention is likely to have triggered a
manipulative argument, it is not a necessary condition for the interpretative effect to take place as the definition makes
clear. Such a definitional framework readily accounts for cases of manipulation by proxy as they were described above.
An analogy could illustrate further the contrast between manipulation and persuasion. The actions of persuasion and
manipulation could be described as two actions in which a speaker tries to get some information through customs.
Persuasion would then correspond to a situation where the speaker tries to get some information into the CEH by getting a
proper visa for it, while manipulation corresponds to an attempt at smuggling some information into CEH (‘‘getting past [the
hearer’s] defences’’ according to Sperber et al., 2010).
Based on this model, two types of strategies are expected to allow a manipulator to achieve the intended effect on the
hearer. She can either choose to increase the accessibility of the context set C in which U is strengthened or weaken the
accessibility of the context set Cˈ in which U is inconsistent and eliminated as shown in Fig. 1.
Fig. 1. Two types of Context Selection Constraint in manipulative discourse: weakening of Cˈ or strengthening of C in the order of accessibility.
Having laid out the theoretical background for a pragmatic analysis of manipulative discourse, a case study of a
prototypical example of manipulative fallacy is proposed that draws on the explanatory potential of the model and
highlights further points of convergence with the relevance theoretic framework.
4. Not so ill-formed manipulative arguments: the case of the ad populum fallacy
Maillat and Oswald (2009, 2011) explore the interface between argumentation theory and cognitive psychology and
argue that CSC allows the analyst to explain manipulative arguments and in particular their effects on the hearer with an
improved resolution and transparency.
There are many different ways to achieve either one of the two manipulative strategies identified in the previous
section. All of them, however, will target the internal structure of the cognitive environment and in that sense they are
expected to be cognitive strategies. This should not be surprising for a pragmatic theory like Relevance Theory which
asserts its own cognitive validity and foundation. As was already pointed out in an earlier quotation, the ‘cognitive submechanisms’ that govern — among other things — interpretative processes are subject to shortcomings (Sperber and
Wilson, 1995:262).
A good illustration of cognitive shortcomings is found in cognitive psychology under the category of cognitive illusions
(see Pohl, 2004 for a detailed overview). Allott and Rubio Fernandez (2002) also looked at the pragmatic implications of
such cognitive shortcomings in relation with the so-called Moses illusion (see Erickson and Mattson, 1981; Reder and
Kusbit, 1991). Pohl (2004:2–4) defines a cognitive illusion as a cognitive phenomenon that (i) deviates from some ‘correct’
10
It is important to point out here that this contradictory subset Cˈ is not challenged by U, it is simply avoided.
196
D. Maillat / Journal of Pragmatics 59 (2013) 190–199
standard, (ii) deviates in a systematic fashion, (iii) escapes conscious control, and, as a result, (iv) is difficult to avoid.
Unsurprisingly, many manipulative strategies exploit such cognitive illusions in order to mislead the cognitive system.11
According to van Eemeren and Houtlosser (2008:140), the ad populum fallacy can be defined as ‘‘the fallacy of
regarding something acceptable because it is considered acceptable by a great many people’’. In other words, the
repetition of U by several sources tends to strengthen U. One does not need to look very long for an example of such
manipulative use as the history of propaganda and advertising is replete with instances of it. Burst advertising campaigns
for example count on the overwhelming presence and repetition of their slogan to strengthen its impact on their target
audience. Similarly, as was already pointed out in an earlier section, the Nazi regime made sure that the German
population was repeatedly exposed to some central claims (e.g. about the alleged inferiority of some minority ethnic,
religious and political groups) in order to strengthen their accessibility within CEH.
From a CSC perspective, the analysis of the ad populum fallacy is relatively straightforward. The repetition of U
guarantees a greater degree of accessibility as U gets strengthened in CEH with each occurrence and as it gets connected
with an increasing number of context sets. In CSC terms, this corresponds to a strengthening-by-repetition constraint on C
which ensures that inconsistent context sets (which would lead to the elimination of U are fewer and weaker) are not
accessed.
As it turns out, there exists ample evidence in the fields of cognitive psychology and psycholinguistics that supports the
cognitive effect (on CEH) described above. Two directions explored by scholars to elucidate the cognitive mechanisms which
underlie strengthening-by-repetition ought to be mentioned here. Both effects enter the category of cognitive illusions.
First, cognitive psychology identified the mere-exposure effect which is defined as ‘‘increased liking for a stimulus that
follows repeated, unreinforced exposure to that stimulus’’ (Bornstein and Craver-Lemley, 2004:231). While the notion of
‘liking’ does not correspond directly to that of accessibility in the proposed model, I want to argue that the observed
increase in the positive connotation of a given contextual assumption coincides with an increased level of accessibility
(see Zajonc, 1968 for a discussion of mere-exposure effect on a lexical level).
A second, more important, cognitive effect, which has been linked to repeated exposure is to be linked to
strengthening-by-repetition, namely the validity effect. Contrary to the mere-exposure effect, this latter type of cognitive
illusion is firmly grounded within language. Hackett Renner (2004:201) describes its impact on a hearer’s assessment of a
given piece of verbally transmitted information.
[I]f information has been heard previously, people are likely to ascribe more truth or validity to it than if they are
hearing it for the first time [. . .] regardless of whether the information was originally believed to be true or false.
That is to say that the more an assumption is repeated the more valid it becomes in CEH. This means that through the
simple repetition of U a CEH in which U was eliminated can be turned into a CEH that is consistent with U. This
phenomenon, it is argued, constitutes the cognitive counterpart of the ad populum fallacy. It provides a cognitive
explanation to the CSC strategy of strengthening-by-repetition. Interestingly, the experimental results obtained in
psycholinguistics show that whereas the ad populum might be an ill-formed argument from a rational point of view, it
constitutes a perfectly valid argument from a cognitive perspective, since the repetition of U indeed guarantees that U is
strengthened in CEH. As a result if everybody says U, U effectively becomes more accessible in CEH.
However a second explanation needs to be put forward if one is to account for the most common form of ad populum
argument used in manipulative discourse. While this type of account would explain why and how propaganda or
advertising use repeated exposure to U in order to increase U’s strength in CEH, it is much less clear why an ad populum
fallacious argument which manipulates a hearer into accepting U by merely stating once that everybody says U would
work at all. Nevertheless, an utterance like everybody says U so it’s got to be true instantiates one of the most common,
everyday forms of manipulative fallacy belonging to the ad populum type.
In order to understand what is happening in the latter case, it should first be pointed out that everybody says U so U is a
summary of what would effectively happen on a cognitive level if U were to be repeated by everybody in the hearer’s
environment. As we saw above, the repetition of U strengthens U in CEH. Thus, to the extent that the antecedent of such
an argumentative utterance describes an actual state of affair (it is true that everybody says U), the consequent cognitively
follows from the mere-exposure and validity effects (U will be vindicated in CEH). Therefore, everybody says U so U must
be true can be regarded as a discursive shortcut for strengthening-by-repetition. Such an argumentative move relies on a
cognitive effect whereby a contextual assumption is reinforced through the mere mention of the corresponding cognitive
process. In that sense, the ad populum only mentions repetition without actually realising it, but the cognitive effect on CEH
remains.
Incidentally, it seems important to point out that ample evidence that such cognitive shortcuts can take place has been
found within the field of cognitive psychology, as scholars addressed the contrast between use and mention of repetition.
11
Maillat and Oswald (2009, 2011) identify other such manipulative illusions.
D. Maillat / Journal of Pragmatics 59 (2013) 190–199
197
The goal of such research is to establish if the simple fact of mentioning that a given stimulus has been repeated (mention
condition) suffices to trigger the validity effect noted when the stimulus in question is actually repeated (use condition). The
team of scholars who published their results as Bacon (1979:247) showed that the triggering factor for the validity effect is
a recognition judgment for a given stimulus more than the actual repetition of the stimulus in question. They write that:
‘Unambiguously, then, differences in rated truth are more sensitive to recognition judgments than to actual repetition of the
statements being rated’. Ozubko and Fugelsang (2011:273) showed further that the same effect can be generated by
misleading subjects in their belief that they encountered a target stimulus before. Coming back to the ad populum, it could
therefore be argued that it exploits the cognitive effect of repetition in mention.
This was the effect sought by Colin Powell in the speech he delivered on 5 February 2003 to the UN Security Council.
He justified then that military operations should be envisaged against Iraq, based on — arguably — numerous pieces of
information:
The material I will present to you comes from a variety of sources. Some are U.S. sources and some are those of
other countries. Some [of] the sources are technical, such as intercepted telephone conversations and photos taken
by satellites. Other sources are people who have risked their lives to let the world know what Saddam Hussein is
really up to.
I cannot tell you everything that we know, but what I can share with you, when combined with what all of us have
learned over the years, is deeply troubling.
In this preamble, Powell tries to convey the idea of a very large number of sources (‘a variety of sources. . . Some are. . .
Some of the sources are. . . Other sources are. . .’) and pieces of evidence that assert U, namely that Iraq is an international
threat. We know now that there were very few and poor sources, but the manipulative move is clearly that of an ad
populum argument which implies that all of us know that everything we know confirms U (‘everything that we know. . . what
all of us have learned over the years’) even though the evidence actually presented is rather scarce.
With these observations, we reach the conclusion that while certain interpretative effects can be achieved by using
language in an appropriate way (by repeating a target utterance for instance), there exists linguistic shortcuts to trigger
those same effects by simply alluding to the conditions under which the effect is generated. Unsurprisingly, such shortcuts
exert a very strong appeal on manipulators. The relevance-theoretic framework imparts the analyst with a powerful tool in
connection with this last observation. Unlike many other pragmatic models, RT has looked at ways that languages have to
control the inferential processes at work during interpretation. In her work on the pragmatics of discourse markers,
Blakemore (2002:89) notes that languages have evolved ‘‘coded means for constraining the inferential tasks involved in
utterance interpretation’’. These expressions are called ‘procedural expressions’. The claim put forward in this paper is
therefore that an utterance such as everybody says U functions like a procedural expression in that it constrains the
inferential system towards strengthening-by-repetition, thereby effectively manipulating the interpretation of U
(strengthening U in CEH). In that sense, everybody says U opens a cognitive shortcut since ‘‘[procedural expressions]
ensure the recovery of the intended cognitive effects for a minimum cost in processing’’ (Blakemore, 2002:130).
5. Conclusion and perspective
In this model, manipulative effects are achieved by constraining the context selection process. In the case in point, the
ad populum fallacies, the constraint can be imposed on CEH in two distinct ways. The manipulator can adopt a brute-force
strategy and decide to coerce interpretation cognitively by increasing the degree of accessibility of C through repetition; or
she can use a procedural expression (everybody says U) to constrain the inferential process and guide it through a
cognitive shortcut.
The existence and appeal of such cognitive shortcuts within the interpretative process is predicted by a pragmatic
theory based on an optimisation procedure like Relevance Theory. In a relevance-theoretic framework, interpretation is
efficiency-driven. From an evolutionary perspective such a system is expected to look for processing shortcuts.
Manipulation takes advantage of this natural — inevitable — attraction towards cognitive shortcuts and exploits it. And in
many cases it gets away with it because, as Sperber et al. (1995:90) underline:
[. . .] people are nearly-incorrigible ‘‘cognitive optimists’’. They take for granted that their spontaneous cognitive
processes are highly reliable, and that the output of these processes does not need re-checking.
In a recent development of these ideas, Sperber et al. (2010) claim that the cognitive system evolved a separate
mechanism, ‘epistemic vigilance’, in order to offset the cognitive drawbacks of this drive towards efficiency. The role of the
epistemic vigilance filter is to ensure that the cognitive mechanisms governing interpretation are not misled or trapped too
often. In that sense epistemic vigilance is predicted to be intimately related to manipulative uses of language, as they are
expected to function in opposite directions.
198
D. Maillat / Journal of Pragmatics 59 (2013) 190–199
The literature on fallacious arguments (see van Eemeren and Grootendorst, 2004) provides ample evidence of this
antagonistic relationship. Sperber et al. (2010) invoke two main orientations for epistemic vigilance: (i) vigilance towards
the content, and (ii) vigilance towards the source of an utterance. As expected in such a system, some manipulative uses
can be categorised as attempts to lure epistemic vigilance on the source. For example, an expression like experts say that
U — an instance of ad verecundiam fallacy — can be analysed as a procedural expression which constrains the
interpretation of U by weakening the accessibility of any subset Cˈ in which U is eliminated, as the source of U gets
contextually strengthened. An ad personam fallacy works in the opposite direction. Interestingly, some fixed formulations
function like procedural expressions to trigger the interpretative effects associated with the ad verecundiam and ad
personam fallacies. Expressions such as experts/scientists/government officials say that U and idiots claim that U are
conventional linguistic signs that point towards inferential shortcuts which strengthen, respectively weaken, the contextual
assumptions associated with U.
Likewise, we find procedural expressions which signal content-related validity to the epistemic vigilance device.
Syllogistic arguments, for instance, are cognitively safe in the sense that they guarantee the soundness of the conclusion
drawn. Unsurprisingly, however, manipulators know that an argument that resembles a syllogism stands very good
chances of evading the epistemic vigilance filter. This cognitive trick is particularly effective because it relies on a
inconspicuous pragmatic phenomenon first discussed in Geis and Zwicky (1971:565), namely conditional perfection (CP).
CP is described as a form of pragmatic enrichment, an ‘‘invited inference’’, which is ‘‘a principle governing the effects that
utterances have — conditionals are understood to be perfected unless the hearer has reason to believe that the converse is
false [. . .]’’. Perfecting, in this instance, is therefore a case of treating a conditional relation as a biconditional one. By way
of illustration, the conditional in (2), as a consequence of CP, tends to pragmatically license the conclusion expressed (Lou
must be here), even though the affirmation of the consequent in the second premise (I just saw Max) does not validate it
from a logical point of view:
(2) You know what they say: ‘If you see Lou somewhere, Max is there too.’ -And I just saw Max. . . So Lou must be here too.
As Geis and Zwicky (1971) argue, CP is a pervasive phenomenon and although it is not conventional in Grice’s sense
(1989), it functions like a generalised conversational implicature which requires special circumstances in order to be
cancelled.12
For this very reason, manipulation appreciates fallacious arguments disguised as syllogisms, such as affirming the
consequent, as they rely on a generalised conversational implicature — Conditional Perfection — which tends to be
triggered by default (in Levinson’s (2000) sense). As a result, a manipulator can try to trick the addressee by counting on
the fact that a default CP implicature tends to be applied to conditionals if the addressee fails to note that the converse
relation does not hold. Thus, while the application of CP in (3) is unproblematic, its application in (4) is.
(3) At the time of each North Korean nuclear test, we observe increased movement around that army base. And
spies spotted intense activity on the base.
[ A North Korean nuclear test is underway. (via CP)
(4) At the time of each North Korean nuclear test, we observe rain in Hamburg. And it rained in Hamburg yesterday.
?? [ A North Korean nuclear test is underway. (common sense calls for a cancellation of CP)
Interestingly for the manipulator, the propensity for the CP implicature to be drawn even in (4) provides her with a
perfect tool to strengthen U — in this case, the conclusion of the syllogism — in the cognitive environment of the
addressee. In this second type of manipulative discourse, the cognitive shortcut is triggered through the content of
the manipulating utterance (as argued in Sperber et al., 2010); when the addressee recognises a conditional
structure, there is a great risk that he will tend to apply the default CP. When he does, he will have been manipulated
to reinforce the conclusion U in his CE. For this same reason, we expect many other manipulative moves and
fallacies to exploit generalised conversational implicatures, as they constitute a form of default cognitive shortcut. In
our example, the quasi-syllogistic structure exploits the inferential constraints triggered by genuine syllogistic
procedural expressions of the form (if P then Q; P; so Q).
While these examples can only hint at the analytic potential of the model, I hope to have established that this type of
pragmatic approach to manipulative uses of language opens up a very rich and fresh perspective on debates which have
occupied language scholars since Aristotle.
12
As the quotation above makes clear, CP is to be cancelled if the hearer ‘‘has reason to believe that the converse is false’’ (ibid.).
D. Maillat / Journal of Pragmatics 59 (2013) 190–199
199
References
Allott, N., Rubio Fernandez, P., 2002. This paper fills a much-needed gap. In: Afuta, P., El Ghali, A., Toussenel, F. (Eds.), Actes de l’Atelier des
doctorants en linguistique. Université Paris 7, Paris, pp. 97–102.
Attardo, S., 1997. Locutionary and perlocutionary cooperation: the perlocutionary cooperative principle. J. Pragmatics 27, 753–779.
Bacon, F.T., 1979. Credibility of repeated statements: memory for trivia. J. Exp. Psychol. Hum. Learn. 5, 241–252.
Blakemore, D., 2002. Relevance and Linguistic Meaning: The Semantics and Pragmatics of Discourse Markers. Cambridge University Press,
Cambridge.
Bornstein, Robert F., Craver-Lemley, Catherine, 2004. Mere exposure effect. In: Pohl, R.F. (Ed.), Cognitive Illusions: A Handbook on Fallacies
and Biases in Thinking, Judgment and Memory. Psychology Press, New York, pp. 215–234.
Carston, R., 2002. Thoughts and Utterances: The Pragmatics of Explicit Communication. Blackwell, Oxford.
Cummings, L., 2005. Pragmatics: A Multidisciplinary Perspective. L. Erlbaum Associates, Mahwah, NJ.
de Saussure, L., 2005. Manipulation and cognitive pragmatics: preliminary hypotheses. In: de Saussure, L., Schulz, P. (Eds.), Manipulation and
Ideologies in the Twentieth Century: Discourse, Language, Mind. Benjamins, Amsterdam, pp. 113–145.
de Saussure, L., Schulz, P. (Eds.), 2005. Manipulation and Ideologies in the Twentieth Century: Discourse, Language, Mind. Benjamins,
Amsterdam.
Erickson, T.A., Mattson, M.E., 1981. From words to meaning: a semantic illusion. J. Verb. Learn. Verb. Behav. 20, 540–552.
Geis, Michael L., Zwicky, Arnold M., 1971. On invited inferences. Linguist. Inq. 2, 561–566.
Goebbels, J., 1938. Signale der neuen Zeit. 25 ausgewählte Reden von Dr. Joseph Goebbels. Zentralverlag der NSDAP, Frz. Eher Nachf.,
Munich, pp. 28–52.
Grice, H.P., 1989. Studies in The Way of Words. Harvard University Press, Cambridge, MA.
Hackett Renner, C., 2004. Validity effect. In: Pohl, R.F. (Ed.), Cognitive Illusions: A Handbook on Fallacies and Biases in Thinking, Judgment and
Memory. Psychology Press, New York, pp. 201–213.
Levinson, S.C., 2000. Presumptive Meanings: The Theory of Generalized Conversational Implicature. MIT Press, Cambridge, MA.
Maillat, D., Oswald, S., 2009. Defining manipulative discourse: the pragmatics of cognitive illusions. Int. Rev. Pragmatics 1, 348–370.
Maillat, D., Oswald, S., 2011. Manipulating contexts: a relevance-theoretic take on manipulation. In: Hart, C. (Ed.), Critical Discourse Studies in
Context and Cognition. John Benjamin, Amsterdam, pp. 65–80.
Murphy, J., 1939. Translation of A. Hitler, Mein Kampf, vol. 1. Hurst & Blackett, London.
Ozubko, J.D., Fugelsang, J., 2011. Remembering makes evidence compelling: retrieval from memory can give rise to the illusion of truth. J. Exp.
Psychol. Learn. Memory Cogn. 37 (1), 270–276.
Parret, H., 1978. Eléments d’une analyse philosophique de la manipulation et du mensonge. Prépublication. Documents de Travail et
Prépublications. Centro Internazionale di Semiotica e di Linguistica. Universita di Urbino Urbino 70, 1–33.
Pohl, R., 2004. Cognitive Illusions: A Handbook on Fallacies and Biases in Thinking, Judgement and Memory. Psychology Press, Hove.
Reder, L.M., Kusbit, G.W., 1991. Locus of the Moses illusion: imperfect encoding, retrieval, or match? J. Mem. Lang. 29, 633–654.
Rigotti, E., 2005. Towards a typology of manipulative processes. In: de Saussure, L., Schulz, P. (Eds.), Manipulation and Ideologies in the
Twentieth Century: Discourse, Language, Mind. Benjamins, Amsterdam, pp. 61–83.
Sperber, D., Wilson, D., 1995. Relevance: Communication and Cognition. Blackwell Publishers, Oxford.
Sperber, D., Cara, F., Girotto, V., 1995. Relevance theory explains the selection task. Cognition 57, 31–95.
Sperber, D., Clément, F., Heintz, C., Mascaro, O., Mercier, H., Origgi, G., Wilson, D., 2010. Epistemic Vigilance. Mind Lang. 25, 359–393.
van Dijk, T., 2006. Discourse and manipulation. Discourse Soc. 17, 359–383.
van Eemeren, F.H., Grootendorst, R., 2004. A Systematic Theory of Argumentation: The Pragma-Dialectical Approach. Cambridge University
Press, Cambridge.
van Eemeren, F., Houtlosser, P., 2008. Rhetoric in a dialectical framework: fallacies as derailments of strategic manoeuvring. In: Weigand, E.
(Ed.), Dialogue and Rhetoric. John Benjamins, Amsterdam, pp. 133–152.
Wilson, D., Sperber, D., 2012. Meaning and Relevance. Cambridge University Press, Cambridge, New York.
Zajonc, R.B., 1968. Attitudinal effects of mere exposure. J. Pers. Soc. Psychol. 9, 1–27.
Didier Maillat received his PhD from Oxford University. He has worked as Associate Professor of English Linguistics at the University of Fribourg,
Switzerland, since 2006. His research focuses on pragmatic models at various interfaces. His early research was on the pragmatics of spatial
reference frames using an experimental framework. More recently, his interest in empirical pragmatics has taken him within the realm of
manipulative discourse. His most recent publications include two theoretical discussions of a pragmatic account of manipulation (2009, 2011; with
Steve Oswald).
U.S. Copyright Law
(title 17 of U.S. code)
governs the reproduction
and redistribution of
copyrighted material.
Downloading this
document for the
purpose of
redistribution is
prohibited.
Eleventh Edition
The Social Animal
Elliot Aronson
University of California, Santa Cruz
with Joshua Aronson
New York University
WORTH PUBLISHERS
To Vera, of course
The Social Animal, Eleventh Edition
Acquisitions Editor: Erik Gilg
Marketing Manager: Jennifer Bilello
Art Director: Babs Reingold
Senior Designer: Kevin Kali
Senior Project Editor: Laura McGinn
Copy Editor: Maria Vlasak
Production Manager: Sarah Segal
Compositor: Northeastern Graphic, Inc.
Printing and Binding: R. R. Donnelley
Cover Drawing by Tom Durfee
Library of Congress Control Number: 2007925852
ISBN-13: 978-1-4292-3341-5
ISBN-10:1-4292-3341-9
© 2012, 2008, 2004,1999 by Worth Publishers
©1995,1992,1988,1984,1980,1976,1972 by
W . H. Freeman and Company
Printed in the United States of America
First printing 2011
Worth Publishers
41 Madison Avenue
New York, NY 10010
www.worthpublishers.com
132 The Social An imal
makes that person seem more similar to others, and thus more average and like everyone else. An average person is less likely to have an
extremely high grade point average or to be terribly negative.
Judgmental Heuristics
One way that we make sense of the buzzing, blooming array of information that comes our way is through the use of judgmental
heuristics. A judgmental heuristic is a mental shortcut; it is a simple, often only approximate, rule or strategy for solving a problem.32
Some examples include “If a man and a woman are walking down a
street, the man walks on the outside.” “If a particular food item is
found in a health food store, it must be good for you.” “If a person is
from a rural town in Arkansas, he or she must be intellectually backward.” Heuristics require very little thought—-just the selection of
the rule (which may not be the correct one to use) and a straightforward application to the issue at hand. It can be contrasted with more
systematic thinking in which we may look at a problem from a number of angles, assemble and evaluate as much relevant information as
possible, and work out in detail the implications of various solutions.
Let’s look at three of the most common judgmental heuristics—the
representative heuristic, the availability heuristic, and the attitude
heuristic.
The Representative Heuristic According to Daniel Kahneman and Amos Tversky,33 when we use the representative heuristic,
we focus on the similarity of one object to another to infer that the
first object acts like the second one. For example, we know that highquality products are expensive; therefore, if something is expensive,
we might infer that it is really good. Thus, if I see two bottles of wine
on the shelf and one has a higher price, I leap to the conclusion that
the more expensive one is the better wine. I select the one feature
(price) from among the many others that I might have focused on—
such as type of grape, vintner, vintage, wine-growing region—and I
use that to make my decision. But, as most smart consumers know,
high price does not always mean high quality. Let’s look in more detail at the implications of the use of the representative heuristic by
eavesdropping on a conversation between mother and child in the
aisle of a local supermarket.
Social Cognition
133
Picture the scene: Seven-year-old Rachel spots her favorite cereal, Lucky Charms, takes a box off the shelf, and quietly delivers it
to the shopping cart. Her mom looks at the box in disgust. It is bright
red. A leprechaun is sprinkling shining stars (must be sugar) over
pink and purple marshmallow bits. On the back of the box, her mom
finds a message informing her that a special, enclosed pair of glasses
can be used to find hidden leprechauns.
Mom sternly announces, “Rachel, put that junk back on the
shelf. It is loaded with sugar and nothing but empty calories.”
Rachel replies, “But, Mom, it tastes good.”
Being a smart mom, she offers Rachel another choice and a little
inducement. “Why not try this one? It’s called 100 Percent Natural
Granola. It is good for you. Eat this and you’ll grow up to be a big girl.”
Rachel looks at the box. It is small but heavy. The picture on the
front features a bowl of light brown cereal set against a wood-grain
background and a couple of stalks of unprocessed grains. On the
back of the box is a lot of small, hard-to-read writing.
Rachel exclaims, “Yukko! I don’t want to be a big girl.”
How would you resolve the great breakfast cereal standoff?
Would you side with the mother and opt for nutrition, even though
Rachel may not like the taste? Or would you feel that Rachel, even
at this tender age, should be making her own decisions, regardless of
the consequences? My recommendation may surprise you: The fight
is for naught. Tell Rachel and her mom to buy the Lucky Charms
because, in actuality, it is more nutritious than the “natural” cereal. If
Rachel’s mom had bothered to read the fine print and conducted a
systematic comparison between Lucky Charms and 100 Percent
Natural Granola, she would have discovered that Lucky Charms is
lower in calories and saturated fats than 100 Percent Natural Granola.34 Although Lucky Charms is also slightly higher in sugar, this
difference is negligible and of little dietary importance. Indeed, in
1981, Consumer Reports, a highly respected source of consumer information, conducted a test of breakfast cereals.35 Their researchers
fed young rats, which have nutritional requirements remarkably similar to those of humans, a diet composed exclusively of water and one
of 32 brands of breakfast cereal for a period of 14 to 18 weeks. They
found that the rats grew and remained healthy on a diet of Lucky
Charms. On the other hand, a diet of Quaker’s 100 Percent Natural
Granola actually stunted their growth!
134 The Social An imal
What caused the disagreement between Rachel and her mom? It
is clear that they used the cereal package (not the cereal) as a representative heuristic. In this case, the problem for Mom was to select a nutritious cereal; for Rachel the problem was to get a cereal that was fun
and tasty. The box of Lucky Charms resembles a child’s toy—bright
colors, cartoon character, glistening sugar. We infer that this cereal is
“childish,” and since children eat junk food if not carefully supervised,
this cereal must be junk. On the other hand, the 100 Percent Natural
Granola box has the earth tones and a picture of unprocessed grains;
it resembles nature itself. And, of course, the brand name is consistent; it is “natural” and, in our minds, the natural is equated with the
good, the wholesome. The cereal must be nutritious.
The representative heuristic is used in places other than the
supermarket.36 An analysis of folk remedies and early Western medicine shows that a common assumption is that the cure should resemble the cause of the disease. For example, in one culture, epilepsy
is treated with a drug made from a monkey whose movements appear epileptic. Similarly, in Western culture, newspapers initially
ridiculed Walter Reed’s suggestion that yellow fever was carried by a
mosquito, since there is little resemblance between the cause (mosquitoes) and the result (yellow fever). The representative heuristic is
used to identify psychological causes, as well. For example, in the
1960s and 1970s, many conservative adults clung to the belief that
the political radicalism exhibited by the college students of that era
was caused by permissive child-rearing practices. In early psychoanalytic theorizing, an obsessive-compulsive personality was known as
anal retentive and was believed to be the direct result of early and severe toilet-training practices. In the overwhelming majority of
American presidential elections, the taller of the two major candidates has emerged victorious—suggesting the possibility that some
Americans may implicitly believe that height may have something to
do with the ability to lead.
The representative heuristic is often used to form impressions
and to make judgments about other persons. The first information
we pick up about a person—information about gender, race, physical attractiveness, and social status—is usually associated with simple rules that guide thought and behavior. Gender and ethnic
stereotypes tell us “just how men and women differ” and “what a particular member of an ethnic group is like.” Much research has
Social Cognition
135
demonstrated that most people leap to the conclusion that beautiful
people are more successful, sensitive, warmer, and of better character
than less attractive people. Persons of high social stature, often inferred by dress and mannerisms, are respected and held in high esteem. Is it any wonder that “get ahead” self-help books often describe
how to take advantage of these heuristics by urging their readers to
“dress for success”; that is, to wear the kinds of clothes that will create the image of a successful person? This is the representative
heuristic in action.
The Availability Heuristic Suppose you go to a restaurant
with some friends. Your friend Neal orders a steak with onion rings,
but the waiter brings his steak with fries instead. “Oh, well,” he says.
“No big deal—I like fries almost as much as onion rings.” This opens
a discussion as to whether he should have sent back his order. Marlene accuses Neal of being unassertive. He turns to you and asks, “Do
you think I’m an unassertive person?” How would you answer this
question?
If you know Neal well and have already formed a picture of how
assertive he is, you can recite your answer easily and quickly. Suppose,
however, that you’ve never really thought about how assertive Neal
is. In this kind of situation, most of us will rely on how quickly and
easily an example might come to mind. If it is easy to think of one
vivid occasion when Neal acted assertively (e.g., “that time he
stopped someone from cutting in line in front of him at the movies”),
you will conclude that Neal is a pretty assertive guy. If it is easier to
think of an occasion when Neal acted unassertively (e.g., “that time
he let a phone solicitor talk him into buying a Slap-Chop for
$29.99”), you will conclude that he is pretty unassertive.
This mental rule of thumb is called the availability heuristic,
which refers to judgments based on how easy it is for us to bring specific examples to mind. There are many situations in which the availability heuristic will prove accurate and useful. Specifically, if you can
easily bring to mind several examples of Neal standing up for his
rights, he probably is an assertive person; if you can easily bring to
mind several examples of Neal letting people push him around, he
probably is not. The main problem with employing the availability
heuristic is that sometimes what is easiest to bring to mind is not
typical of the overall picture. This will lead us to faulty conclusions.
136 The Social An imal
Let’s try something: Do you think more people in the United
States die from shark attacks or from falling airplane parts? Do you
think more people die from fires or from drowning? Think about it
for a minute.
When asked those questions, the overwhelming majority of people report that deaths from shark attacks are more common than those
from falling airplane parts and that deaths from fires are more common than those from drowning. In fact, both answers are wrong. Why
do most people believe these things? Studies suggest that is probably
easier to bring to mind examples of deaths from sharks and fires because these events are more likely to be covered in a vivid manner on
the evening news and thus are more available in people’s memories.37
Similarly, if you ask people to estimate the number of violent
crimes committed each year in the United States, you will get very
different answers, depending on how much prime-time television
they watch, as we learned in Chapter 3. People who watch a great
deal of television—and, hence, see a great deal of fictionalized
violence—vastly overestimate the amount of real crime that occurs in
our nation.38
The Attitude Heuristic An attitude is a special type of belief
that includes emotional and evaluative components; in a sense, an attitude is a stored evaluation—good or bad—of an object. According
to Anthony Pratkanis and Anthony Greenwald,39 people tend to use
the attitude heuristic as a way of making decisions and solving problems. Attitudes can be used to assign objects to a favorable class (for
which strategies of favoring, approaching, praising, cherishing, and
protecting are appropriate) or to an unfavorable category (for which
strategies of disfavoring, avoiding, blaming, neglecting, and harming
are used). For example, if John dislikes President Obama, he blames
his policies for the recession, the high level of unemployment, and
the huge deficit; if he likes Obama, he is apt to blame these problems on Obama’s predecessor.
Much research has shown that attitudes can be used to make
sense of our social world. For example, a study by Anthony Pratkanis40 found that a person’s attitudes play a major role in determining
what he or she “knows” to be true. In this study, college students were
asked to identify which of two possible statements about another former president—such as the following—was true:
Social Cognition
137
a. Ronald Reagan maintained an A average at Eureka College.
b. Ronald Reagan never achieved above a C average at Eureka
College.
What did Pratkanis find? Very few people actually knew what
Reagan’s college grades were; their answer depended on their attitude
toward him. Students who liked Reagan were more likely to believe
statement (a); students who disliked him were more likely to believe
statement (b). What is more, the more extreme the attitude toward
Reagan, the more confidence the students had in their judgments. In
other words, the students in this study used their attitudes to estimate reality and then believed that what they estimated was correct.
For those of you who are curious, statement (b) is correct. Reagan
never achieved above a C average in college. (I hasten to add that this
is an actual fact and has nothing to do with my personal attitude toward Mr. Reagan!)
The use of an attitude heuristic can influence our logic and ability to reason. For example, in the late 1940s, Donald Thistlewaite41
asked respondents to state whether syllogisms such as the following
were valid:
Premise 1: If production is important, then peaceful industrial relations are desirable.
Premise 2: If production is important, then it is a mistake to have
Negroes for foremen and leaders over Whites.
Therefore: If peaceful industrial relations are desirable, then it is a
mistake to have Negroes for foremen and leaders over Whites.
A moment’s reflection shows that the syllogism, as stated, is fallacious; the conclusion does not logically follow from the premises.
But Thistlewaite found that prejudiced individuals (who agree with
the conclusion) are far more likely to indicate (incorrectly) that the
logic is valid than are less prejudiced people.
Another dimension of the attitude heuristic is the halo effect, a
general bias in which a favorable or unfavorable general impression
of a person affects our inferences and future expectations about that
person. For example, if you really like President Obama, then you
will be likely to discount or explain away any behavior on his part
that might be considered negative, and exaggerate the goodness of
138 The Social An imal
his positive actions. In your mind, it is almost as if he is wearing an
angel’s halo. Similarly, a disliked individual is assumed to possess
negative traits, with his or her performance subsequently devalued.
In one experiment, Richard Stein and Carol Nemeroff42 demonstrated that college students gave a halo (both positive and negative)
to women, depending upon the kinds of food they ate: All other
things being equal, once they found out that a woman ate health
food, they rated her as more feminine, more physically attractive, and
more likable than junk food eaters.
Still another dimension of the attitude heuristic is the falseconsensus effect. Almost all of us have a tendency to overestimate
the percentage of people who agree with us on any issue. If I believe
something, then I will tend to assume that most other people feel the
same way. For example, in one experiment, Lee Ross and his colleagues43 asked college students if they were willing to wear a sign
around the campus that said “Eat at Joe’s.”Those who agreed to wear
the sign thought that most other people would, too; those who decided against wearing the sign estimated that few other students
would wear it. In other words, we often make the (not necessarily
true) assumption that others like what we like and do what we prefer to do.
When Do We Use Heuristics? Of course, decisions don’t have
to be based on heuristics. Rachel’s mother might have carefully read
the ingredients on the cereal box, subscribed to a consumer magazine, or consulted nutrition textbooks. Similarly, we could carefully
reason about an issue or study the record and accomplishments of a
politician; this would make us less likely to use our attitudes as a simple way to make sense of the world. And, occasionally, most of us do
go through the decision-making process in a rational manner.
This raises an important question: What conditions are most
likely to lead to heuristic employment rather than rational decision
making? Research has identified at least five such conditions.44 As
you might expect from our earlier discussion of humans as cognitive misers, heuristics are most likely to be used when we don’t have
time to think carefully about an issue, or when we are so overloaded
with information that it becomes impossible to process the information fully, or when the issues at stake are not very important, so
that we do not care to think about it. Heuristics are also used when
Social Cognition
139
we have little solid knowledge or information to use in making a
decision.
A moment’s thought will reveal that the persuasion landscape
faced by Rachel and her mother contains many of the features that
lead to heuristic decision making. If she is like most Americans,
Rachel’s mother is feeling increasingly time-pressed since her leisure time has eroded considerably in the last 10 years. As a consumer, she faces a message-dense environment complete with a
choice of nearly 400 different brands of cereal currently on the market. She probably has had little consumer education or training. At
the same time, she has been the recipient of millions of advertisements, each repeating and repeating a brand image, so that this
image will quickly come to mind in the aisles of the local supermarket. Given this state of affairs, it is a wonder that all decisions aren’t
made heuristically.
Categorization and Social Stereotypes
Before the Persian Gulf War of 1991, the U.S. Congress held a series of debates on the positive and negative consequences of going to
war. Those who supported the war described Saddam Hussein as the
“new Hitler”; they emphasized the parallels between Saddam’s
gassing of the Kurds and Hitler’s gassing of the Jews, Iraq’s invasion
of Kuwait and Germany’s invasion of Poland and the Baltics, and
Saddam’s and Hitler’s buildup of armaments. Those who opposed
the war saw the situation in Iraq as paralleling that of Vietnam; they
saw both incidents as civil wars—a fight between North and South
Vietnam and between various Arab factions; they worried about the
U.S. military’s ability to fight in foreign terrain of swamps and
deserts; they characterized the war efforts as a war in support of “big
business” and “big oil.”
In a sense, the debate over whether to go to war with Iraq was really a debate over whose categorization of ambiguous events was correct. And with good reason. For once it is decided how an event or
person should be categorized, it becomes clear what course of action
should be taken. If Saddam is truly a “new Hitler,” then the policy of
economic sanctions (which some considered a form of appeasement)
will only bring additional threats to peace and ultimately a much
worse war. If Iraq is another Vietnam, then intervention would lead
Chapter 4 451
29. Jones, E. E., Rock, L., Shaver, K. G., Goethals, G. R., & Ward, L. M . (1968). Pattern
of performance and ability attribution: An unexpected primacy effect .Journal of Personality and Social Psychology, 10, 317-340.
30. Aronson, J. M., & Jones, E. E. (1992). Inferring abilities after influencing performances. Journal of Experimental Social Psychology, 28, 277-299.
31. Zukier, H. (1982). The dilution effect: The role of the correlation and the dispersion of
predictor variables in the use of nondiagnostic information. Journal of Personality and
Social Psychology, 43, 1163-1174.
See also: Denhaerinck, P., Leyens, J., & Yzerbyt, V. (1989). The dilution effect and
group membership: An instance of the pervasive impact of outgroup homogeniety. European Journal of Social Psychology, 19, 243-250.
32. Nisbett, R., & Ross, L. (1980). Human inference: Strategies and shortcomings of social
judgment. Englewood Cliffs, NJ: Prentice-Hall.
Sherman, S. J., & Corty, E. (1984). Cognitive heuristics. In R. S. Wyer & T . K. Srull
(Eds.), Handbook of social cognition (Vol. 1, pp. 189-286). Hillsdale, NJ: Erlbaum.
Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and
biases. Science, 185, 1124-1131.
33. Kahneman, D., & Tversky, A. (1973). On the psychology of prediction. Psychological
view, 80, 237-251.
Re-
34. Comparison based on U.S. Department of Health and Human Services. (1987). Eating to lower your high blood cholesterol (NIH Pub. No. 87-2920). Saturated fats (such as
animal fat) and some vegetable fats (such as coconut oil, cocoa butter, palm oil, and hydrogenated oils) have been shown to raise cholesterol levels. Cereal manufacturers
sometimes use such fats, especially coconut, palm, and hydrogenated oils, in their products. Many manufacturers are in the process of reformulating their cereals to remove
such oils, so check the label before making a final decision.
35. “Which cereal for breakfast?” (1981, February). Consumer Reports, pp. 68-75.
36. Nisbett, R., & Ross, L. (1980). Human inference: Strategies and shortcomings of social
judgment. Englewood Cliffs, NJ: Prentice-Hall.
Shweder, R. (1977). Likeness and likelihood in everyday thought: Magical thinking
in judgments about personality. Current Anthropology, 18, 637-658.
37. Pious, S. (1993). The psychology of judgment and decision making. New York: McGrawHill.
Manis, M., Shedler, J., Jonides, J., & Nelson, T. E. (1993). Availability heuristic in
judgments of set size and frequency of occurrence. Journal of Personality and Social Psychology, 65, 448-457.
Schwarz, N., Bless, H., Strack, E, Klumpp, G., Rittenauer-Schatka, H., & Simmons,
A. (1991). Ease of retrieval as information: Another look at the availability heuristic.
Journal of Personality and Social Psychology, 61, 195-202.
Tversky, A., & Kahneman, D. (1973). Availability: A heuristic forjudging frequency
and probability. Cognitive Psychology, 5, 207-232.
38. Signorielli, N., Gerbner, G., & Morgan, M . (1995). Violence on television: The Cultural Indicators Project. Journal of Broadcasting and Electronic Media, 39, 278-283.
39. Pratkanis, A. R. (1989). The cognitive representation of attitudes. In A. R. Pratkanis,
12
The Social An imal
S. J. Breckler, & A. G. Greenwald (Eds.) Attitude structure andfunction
(pp. 71-98).
Hillsdale, NJ: Erlbaum.
Pratkanis, A. R., & Greenwald, A. G. (1989). A socio-cognitive model of attitude
structure and function. In L. Berkowitz (Ed.), Advances in experimental social psychology
(Vol. 22, pp. 245-285). New York: Academic Press.
40. Pratkanis, A. R. (1988). The attitude heuristic and selective fact identification. British
Journal of Social Psychology, 27, 257-263.
41. Thistlewaite, D. (1950). Attitude and structure as factors in the distortion of reasoning.
Journal of Abnormal and Social Psychology, 45, 442-458.
42. Stein, R. I., & Nemeroff, C. J. (1995). Moral overtones of food: Judgments of others
based on what they eat. Personality and Social Psychology Bulletin, 21(5), 480-490.
43. Ross, L., Greene, D., & House, P. (1977). The “false-consensus effect”: An egocentric
bias in social perception and attribution process .Journal of Experimental Social Psychology, 13, 279-301.
44. Pratkanis, A. R. (1989). The cognitive representation of attitudes. In A. R. Pratkanis,
S. J. Breckler, & A. G. Greenwald (Eds.), Attitude structure andfunction
(pp. 71-98).
Hillsdale, NJ: Erlbaum.
45. For a laboratory demonstration, see Gilovich, T. (1981). Seeing the past in the present:
The effect of associations to familiar judgments and decisions. Journal of Personality and
Social Psychology, 40, 797-808.
46. Darley, J. M., & Gross, P. H. (1983). A hypothesis-confirming bias in labeling effects.
Journal of Personality and Social Psychology, 44, 20-33.
47. Rosenthal, R., &Jacobson, L. (1968). Pygmalion in the classroom. New York: Holt, Rinehart & Winston.
Rosenthal, R. (2002). The Pygmalion effect and its mediating mechanisms. In J.
Aronson (Ed.), Improving academic achievement: Impact of psychologicalfactors
on education. San Diego: Academic Press.
48. Hamilton, D. L., 8c Rose, T. L. (1980). Illusory correlation and the maintenance of
stereotypic beliefs. Journal of Personality and Social Psychology, 39, 832-845.
Hamilton, D., Dugan, P., &Trolier, T. (1985). The formation of stereotypic beliefs:
Further evidence for distinctiveness-based illusory correlations. Journal of Personality
and Social Psychology, 48, 5-17.
49. A. R. Pratkanis (1992) Personal communication.
50. Chapman, L. J. (1967). Illusory correlation in observational report .Journal of Verbal
Learning and Verbal Behavior, 6, 151-155.
Chapman, L. J., & Chapman, J. P. (1967). Genesis of popular but erroneous psychodiagnostic observations. Journal of Abnormal Psychology, 72, 193-204.
Chapman, L. J., & Chapman, J. P. (1969). Illusory correlation as an obstacle to the
use of valid psychodiagnostic signs .Journal of Abnormal Psychology, 74, 271-280.
51. Park, B., & Rothbart, M . (1982). Perception of out-group homogeneity and levels of
social categorization: Memory for the subordinate attributes of ingroup and outgroup
members. Journal of Personality and Social Psychology, 42, 1051-1068.
52. For reviews and discussion, see Tajfel, H. (1981). Human groups and social
Cambridge: Cambridge University Press.
categories.
U.S. Copyright Law
(title 17 of U.S. code)
governs the reproduction
and redistribution of
copyrighted material.
Downloading this
document for the
purpose of
redistribution is
prohibited.
FOURTH
EDITION
PERSUASION, SOCIAL INFLUENCE,
AND COMPLIANCE GAINING
ROBERT H. CASS
California State University, Fullerton
JOHN S. SEITER
Utah State University
Allyn & Bacon
Boston Columbus Indianapolis New York San Francisco Upper Saddle River Amsterdam
Cape Town Dubai London Madrid Milan Munich Paris Montreal Toronto Delhi
Mexico City Säo Paulo Sydney Hong Kong Seoul Singapore Taipei Tokyo
Editor-in-Chief: Karon Bowers
Acquisitions Editor: Jeanne Zalesky
Assistant Editor: Megan Lentz
Senior Managing Editor: Linda Mihatov Behrens
Associate Managing Editor: Bayani Mendoza de Leon
Development Manager: David B. Kear
Associate Developmental Editor: Patrick Barb
Marketing Manager: Blair Tuckman
Media Producer: Megan Higginbotham
Editorial Production Services: GGS Higher Education Resources, A division of PreMedia Global Inc.
Manufacturing Buyer: Mary Ann Gloriande
Cover Manager: Anne Bonanno Nieglos
Cover Image: Michael Blann/Digital Vision/Getty Images
Copyright © 2011, 2007, 2003 Pearson Education, Inc.
All rights reserved. No part of the material protected by this copyright notice may be reproduced or utilized in
any form or by any means, electronic or mechanical, including photocopying, recording, or by any information
storage and retrieval system, without written permission from the copyright owner.
Library of Congress Cataloging-in-Publication Data
Gass, Robert H.
Persuasion, social influence, and compliance gaining / Robert H. Gass, John S. Seiter.—4th ed.
p. cm.
Includes bibliographical references and index.
ISBN-13: 978-0-205-69818-9
ISBN-10: 0-205-69818-2
1. Persuasion (Psychology) 2. Influence (Psychology) 3. Manipulative behavior. I. Seiter, John S.
II. Title.
BF637.P4G34 2010
303.3’42—dc22
2009040887
To my high school sweetheart and wife, Susan, thanks for all your
love and support through the best and worst of times.
—Robert Gass
To Miss Gordon, my second-grade teacher, for knowing that
self-concept is the proper starting place.
—John Seiter
Allyn& Bacon
is an imprint of
PEARSON
To our families—Susan, Jordan, Graham, Debora, Dean, Andy, and
Christian—for doing without us when we were writing and
for putting up with us when we weren’t.
—Robert Gass and John Seiter
10 9 8 7 6 5 4 3 2 1 HPC 13 12 11 10
www.peai5onhigheied.com
ISBN-13: TVA-O-EDS-kTAlA-^
ISBN – 1 0 :
• -205-1^616-2
CHAPTER
What Constitutes Persuasion?
at is persuasion? How broad or narrow is the concept? Is persuasion a subset of
human communication in general, much like baseball is a subset of sports? Or is persuasion an element found in all human communication in the same way that coordination
plays a role in every sport? Not surprisingly, different authors view the concept of persuasion in different ways and have, therefore, adopted different definitions of the term. In this
chapter we explore some of the ways persuasion has been defined. We offer our own rather
broad-based, far-reaching conceptualization of persuasion based on five limiting criteria.
We also offer our own model of what persuasion is (Gass & Seiter, 1997, 2000, 2004) and
examine two additional models (Chaiken, 1979,1980,1987; Eagly & Chaiken, 1993; Petty
& Cacioppo, 1986a, 1986b) of how persuasion functions.
You may have encountered some unusual uses of the term persuasion. For example, we have a friend in the construction industry who refers to his sledgehammer as his
“persuader.” He tends to err on the side of cutting a 2 X 4 board too long, rather than too
short, and then “persuading” it into place. As another example, you may recall seeing
one of those old gangster movies in which a mob boss orders his henchman to take
somebody out back “for a little gentle persuasion,” meaning a beating. Although we
don’t normally associate persuasion with pounding lumber or pummeling people, even
in ordinary usage the term does have a wide variety of meanings. Consider each of the
hypothetical situations in Box 2.1, “What Constitutes Persuasion?” Which of these scenarios do you consider to be persuasion?
Adding to the difficulty of defining persuasion is the fact that persuasion also
goes by a variety of other names. Some of its aliases include terms such as advising,
brainwashing, coercion, compliance gaining, convincing, education, indoctrination,
influence, manipulation, and propaganda. Of course, whether these terms are considered pseudonyms for persuasion, or simply related terms, depends on one’s definition
of persuasion.
Defining a concept is analogous to building a fence. A fence is designed to keep
some things in and other things out. In the same way, a definition encompasses some elements or aspects of a concept within its domain while excluding others. Which “species”
of human communication is to be found inside the “barnyard” of persuasion depends on
the size and shape of the fence a particular author builds. Fortunately, the differences in
various definitions can be clarified, if not resolved, by focusing on two key considerations.
We turn to these next.
22
23
What Constitutes Persuasion?
What Constitutes Persuasion?
1. Muffin notices a grubby-looking weirdo in
one of the front seats of the bus she is
boarding. She opts for a seat toward the rear
of the bus. Did the man “persuade” her to sit
elsewhere?
2. Benny Bigot is the principal speaker at a park
rally to recruit more members to the American
Nazi party. Many of the people who hear
Benny are so turned off by his speech that
they are more anti-Nazi than they were before
they attended the rally. Did Benny “persuade”
them?
3. During a dramatic pause in his lecture for his
3-hour night class, Professor Hohum hears a
student’s stomach growling. The professor
then decides it would be a good time for the
class to take a break. Did the student “persuade” Professor Hohum?
4. Babbs is standing at a street corner, watching
passersby. The first three people she sees are
wearing sweatshirts with political and/or
social slogans emblazoned across the front.
The fourth person to pass by is wearing a
plain white T-shirt. Are the first three people
“persuading” Babbs? Is the fourth?
5. Fifi is contemplating going on a major diet.
She realizes she is overweight because she
BOX 2.1
tips the scales at just under 250 pounds, and
her obesity affects her self-esteem. However,
she has read that obese people who lose lots
of weight typically gain the weight back
within a short period of time and that people
are genetically predisposed to be a certain
weight. She convinces herself that there is no
point in dieting. Did Fifi “persuade” herself?
Bubba is at the supermarket, pondering which
of two brands of beer to purchase, a coldfiltered brew or a fire-brewed brew. After
studying both brands attentively, he opts for
the cold-filtered variety. Unbeknownst to him,
another shopper observed his deliberations.
That shopper then walks over to the display
and selects the same brand. Did “persuasion”
take place?
Trudy is an impressionable freshperson who
is in a jam. She has just realized a term paper
is due in her philosophy class. Desperate,
she asks Rex, who is the captain of the
debate squad, if he will help her. Rex offers
to give her an “A” paper he submitted when
he had the same class two years prior if
Trudy will sleep with him. Is Rex using
“persuasion”?
PURE VERSUS BORDERLINE CASES OF PERSUASION
The first consideration is whether one is interested in pure persuasion, or borderline cases of
persuasion. By pure persuasion, we mean clear-cut cases of persuasion, on which most people would agree. Everyone would agree that a presidential debate, or a television commercial, or an attorney’s closing remarks to a jury are instances of persuasion. Such examples
represent “paradigm cases” (O’Keefe, 1990; Simons, 1986) of persuasion because they are
at the core of what we think of when we envision persuasion at work. Other instances,
though, lie closer to the boundary or periphery of what we normally think of as persuasion.
These instances we refer to as borderline cases of persuasion. Not everyone would agree
that a derelict’s mere appearance “persuades” passersby to keep their distance. Nor would
everyone agree that involuntary reflexes such as burps, blinking, and pupil dilation constitute “persuasive” phenomena. These cases are less clear-cut, more “iffy.” Much of the
disparity in definitions is rooted in the fact that some authors are concerned with pure persuasion, whereas other authors are concerned with borderline cases as well. It isn’t so much
5 CHAPTER 2
FIGURE 2.1 Preliminary Model of Persuasion.
a matter of being right or wrong as it is a matter of how wide a net each author wishes to
cast. The preliminary model of the scope of persuasion (Figure 2.1) illustrates this distinction in approaches. As the shading in the model suggests, the dividing line between pure
and borderline persuasion is fuzzy, rather than distinct.
Although we don’t think there is a single, correct definition of persuasion, we do
think there are some things that a functional, contemporary definition of persuasion ought
to do. A contemporary definition should take into account the rich complex of verbal,
nonverbal, and contextual cues found in interpersonal encounters, which, as we noted in
Chapter 1, is the arena in which most influence attempts occur. A contemporary definition
also should acknowledge the many subtle, implicit cues that accompany face-to-face
influence attempts. By implicit cues, we mean communication that occurs at a very low
level of awareness, or even unconsciously. As an example, cultural factors might influence a person’s choice of compliance-gaining strategies, without the person even realizing it (Wiseman et al., 1995). Such implicit communication is, in fact, quite common
(Langer, 1978, 1989a, 1989b; Roloff, 1980) and an important ingredient in persuasion.
The definition and model of persuasion that we offer later in this chapter take these features into account.
1
LIMITING CRITERIA FOR DEFINING PERSUASION
A second consideration in defining persuasion involves the limiting criteria that form the
basis for a given definition. Different scholars apply different litmus tests when defining
persuasion. Five basic criteria can be gleaned from the various definitions offered in the
literature (Gass & Seiter, 2004). We examine each of these criteria in turn.
Intentionality
Is persuasion necessarily conscious or purposeful? Is there such a thing as “accidental”
persuasion? Many who write about persuasion adopt a source-centered view by focusing
What Constitutes Persuasion?
25
on the sender’s intent as a defining feature of persuasion. Bettinghaus and Cody (1994)
adopt this view, stressing that “persuasion involves a conscious effort at influencing the
thoughts or actions of a receiver” (p. 5). For some authors, intentionality is the litmus test
that distinguishes persuasion from social influence (Gass & Seiter, 2000, 2004).
Certainly, pure persuasion would seem to be intentional. When we think of obvious
cases of persuasion we tend to think of situations in which one person purposefully tries
to influence another. But what about borderline cases of persuasion? We believe that
many influence attempts take place without any conscious awareness on the part of the
persuader.
As just one instance, parents quite commonly instill beliefs, impart values, and
model behavior for their children, a phenomenon known as social modeling (Bandura,
1977). Yet they may not realize how much of what they say and do is absorbed by their
young-uns. As any parent will attest, many of the lessons parents “teach” their children are
completely unintended. Another form of unintentional influence involves socialization
processes. From the moment children are born, they are socialized into their respective
gender roles, cultural customs, religious practices, and socio-economic habits. Some
socialization processes are mindful, but many are not. A second way in which an intent criterion is problematic is that people do not always know what specific outcome they are
seeking. Face-to-face encounters, in particular, are laden with spontaneity. Social influence
may arise in and through our interaction with others, rather than as a result of planning and
forethought. Sometimes persuasion just happens.
A third problem with relying on an intent criterion involves situations in which there
are unintended receivers. Imagine a scenario in which two people are discussing which
bets to place on a horse race. One tells the other about an inside tip on a horse that’s a “sure
thing.” A third party overhears the conversation and places a wager on the horse. In such
situations, persuaders don’t intend for third parties to be influenced, yet they often are.
Two studies (Greenberg & Pyszczynski, 1985; Kirkland, Greenberg, & Pyszczynski, 1987)
clearly demonstrate the operation of the unintended receiver effect. In these studies, the
researchers created a situation in which third parties overheard an ethnic slur directed
against an African American. The results of both studies revealed that the overheard ethnic
slur led to lower evaluations by the third parties of the individual at whom the slur was
directed. Notice that a reliance on an intent standard for defining persuasion tends to make
senders less accountable for the consequences of their unintended communication. If a
message has harmful effects, the source can disavow any responsibility by claiming “that’s
not what I intended.”
A fourth limitation lies in the difficulty of ascertaining another’s intent. There can be
a difference between a persuader’s stated intent versus his or her actual intent. Who makes
the determination in such cases? The sender? The receiver? A third party? There are many
vagaries involved in determining whose perception counts. An athlete whose conduct on or
off the field is questioned may proclaim, “I’m not a role model,” but fans and the media
may reply, “Oh yes you are.”
Finally, resolving the issue of intent is particularly difficult in interpersonal contexts,
in which both parties may be simultaneously engaged in attempts at influence. When there
are two interactants, whose intent counts? Intent-based definitions, we believe, are illsuited to modern conceptualizations of human interaction as a two-way venture. The linear
7 CHAPTER 2
view of persuasion that such definitions imply, from sender to receiver, ignores opportunities for mutual influence.
Effects
The effects criterion poses the question: Has persuasion taken place if no one is actually
persuaded? Some authors adopt a receiver-oriented definition of persuasion by restricting
its use to situations in which receivers are somehow changed, altered, or affected. Daniel
O’Keefe (1990) underscores this perspective when he writes:
The notion of success is embedded in the concept of persuasion. Notice for instance, that it
doesn’t make sense to say, “I persuaded him, but failed.” One can say, “I tried to persuade
him, but failed,” but to say simply, “I persuaded him” is to imply a successful attempt to
influence, (p. 15)
The stronger version of this perspective views persuasion as successful if it achieves
the specific outcome sought by the persuader. The weaker version of this perspective settles for outcomes falling short of what the persuader ideally had in mind. Although we recognize the attraction of this point of view, we believe there are problems with limiting the
definition of persuasion in this way. We take the position that even if a person is communicating badly, he or she is still communicating. Similarly, we believe that a person can be
engaged in persuasion even if it is ineffective persuasion. The same can be said for most
other activities. A salesperson might fail to close a deal but would still be engaged in selling. A dancer might dance badly, stepping on his or her partner’s toes, but would still be
engaged in dancing. In short, a person can be engaged in an activity whether the person is
doing it well or not.
An effects criterion emphasizes persuasion as a product. But such an orientation
bears little fidelity to current conceptualizations of human communication as a process. If
we think of persuasion only as an outcome or a thing, then an effects orientation makes
perfectly good sense. We maintain that persuasion is better understood as an activity in
which people engage. This is more than semantic quibbling. By approaching persuasion as
a process, scholars and researchers are more likely to gain insights into how it functions, or
what makes it tick, because they are focusing on what’s going on, not simply on how
things turn out.
A second weakness is the same as that already associated with an intent criterion: An
effects criterion embodies a linear view of persuasion, from source to receiver. In faceto-face encounters, however, there isn’t simply a source and a receiver. Both parties may
be simultaneously engaged in persuasion.
A third problem with relying on an effects criterion is that it is often difficult, if not
impossible, to measure persuasive effects. Rotzoll and Haefner (1996), for example, concluded that only 20 to 40 percent of advertising is effective. The other 60 to 80 percent is
also persuasion; it’s just ineffective persuasion. In fact, the ability to measure persuasive
outcomes may hinge entirely on the sensitivity of one’s measuring instruments (scales, surveys, sales figures, etc.). Furthermore, what constitutes the threshold for a successful versus unsuccessful attempt at persuasion? How much attitude or behavior change must take
What Constitutes Persuasion?
27
place to say persuasion has occurred? And what about the occasional odd circumstance in
which persuasion “boomerangs”—that is, a persuader achieves an effect that is contrary to
his or her intended purpose? Such questions, we believe, point out the many vagaries
inherent in relying on an effects criterion.
We do agree that, as with an intent criterion, pure cases of persuasion can usually
be evaluated by their overall effectiveness. Even then, persuasion is rarely an all or nothing venture. If one also wishes to focus on borderline cases of persuasion, one must
accept the fact that partial persuasion is more the rule than the exception. Notice, too,
that there is some tension between relying on intent and effects as limiting criteria: What
is achieved isn’t always what is intended, and what is intended isn’t always what is
achieved. We happen to think some of the most interesting persuasive campaigns are
those that are unsuccessful, or only partially successful, or that in fact, achieve the opposite of the effect being sought.
Free Will and Conscious Awareness
Many authors endorse the view that there is a distinction between persuasion and coercion. This view is…