Shanghai International Studies University
Interlanguage Pragmatics
Rod Ellis
Topic 2: Pragmatic Failure
________________________________________________________________
A. Introduction
We can distinguish two ways of examining illocutionary acts in SLA:
1. Using error analysis to focus on learners’ failure to perform acts in native-like
ways.
2. Viewing illocutionary acts as part of communicative competence.
B. Pragmatic failure
Thomas (1983) distinguished:
1. Sociopragmatic failure - this occurs when a learner fails to perform the
illocutionary act required by the situation (i.e. deviates with regard to
appropriateness of meaning).
e.g. how L2 learners typically respond to compliments (Wolfson 1989).
Wolfson argues that compliments are used by native speakers of American
English as a means of establishing and maintaining solidarity. It is for this
reason that they are most common among status-equal acquaintances and coworkers rather than among intimates; the former involve more uncertain
relationships which have to be negotiated. Compliments serve as one of the
ways in which Americans (particularly women) undertake this negotiation.
Wolfson points out that many negotiating sequences involving native speakers
are long and elaborate. In comparison, those involving non-native speakers are
typically short, because learners often fail to take up a compliment, preferring
instead to give no response at all:
NS: You have such a lovely accent.
NNS: (No response)
Wolfson argues that by failing to conform to native-speaker complimenting
norms, learners deprive themselves of the opportunities to establish
relationships with native speakers and, thereby, of the input that they need to
develop both their linguistic and sociolinguistic competence.
2. Pragmalinguistic failure – this occurs when a learner tries to perform the right
1
speech act but uses the wrong linguistic means (i.e. deviates with regard to
appropriateness of form).
Not all of Wolfson’s learners manifested sociopragmatic failure through
failing to respond to a compliment. Many displayed pragmalinguistic failure—
that is, they responded to a native-speaker compliment but in linguistically
inappropriate ways. Middle-class, white Americans are likely to respond by
giving unfavourable comments about the object that is the target of a
compliment:
NS: I like your sweater.
NS: It’s so old. My sister bought it for me from Italy a long time ago.
In this way, learners often tried to refuse the compliment or to downgrade
themselves. They also tended to respond with a simple ‘thank you’. Such
responses also served to act as dampeners on the conversation.
Pragmalinguistic failure by learners is widely reported in the literature.
Another good example comes from Eisenstein and Bodman’s (1986) study of
expressions of gratitude. This made use of a discourse completion
questionnaire administered to 67 learners with different L1 backgrounds.
Baseline data were collected from native speakers of English. Eisenstein and
Bodman report that the learners performed very differently from the native
speakers—in fact, their responses were similar only 30 per cent of the time.
They experienced difficulty with both syntax and vocabulary, and also in
identifying the formulas and conventionalized routines that characterized
native-speaker thanking. In a subsequent study (Bodman and Eisenstein 1988),
learners were asked to role play situations calling for expressions of gratitude.
Lower-proficiency learners often translated expressions from their L1. For
example, in thanking someone for a loan they might say ‘May God increase
your bounty’. More advanced learners avoided this kind of pragmalinguistic
error, but instead displayed considerable hesitation and awkwardness.
Pragmalinguistic difficulty, therefore, can also be manifested in the failure to
conform to the temporal norms of native-speaker speech.
The distinction between sociopragmatic and pragmalinguistic errors, however, is not
as clear-cut as these examples may have suggested. Kasper points out that ‘the
distinction becomes fuzzy in the case of indirectness’ (1992: 210). For example, the
decision regarding whether to provide an explanation for having committed some
offence can be seen as a sociopragmatic one, but if providing an explanation is seen as
one of several possible strategies for performing the act of apologizing, its inclusion
or omission constitutes more of a pragmalinguistic decision. Nevertheless, Kasper
acknowledges that the distinction is ‘analytically useful’.
2
C. Illocutionary acts as part of communicative competence
The alternative view of illocutionary acts involves viewing them in terms of
knowledge rather than failure.
Seen in this way the ability to perform illocutionary acts constitutes part of
communicative competence. Canale included this ability in sociolinguistic
competence, which he defined as ‘the extent to which utterances are produced and
understood appropriately in different sociolinguistic contexts’ (1983: 7). He went on
to point out that appropriateness involves both appropriateness of meaning (i.e. when
it is proper to perform a particular illocutionary act) and appropriateness of form (i.e.
the extent to which a given act is realized in a verbal or non-verbal form proper for a
given situation). The former constitutes sociopragmatic competence and the latter
pragmalinguistic competence.
D. The limited scope of research on L2 illocutionary acts
The scope of research in L2 speech acts is quite limited:
1. It has tended to make use of rough and ready categories of sociocultural
reality—the problem of norms referred to earlier. It has also tended to
concentrate on a fairly small set of speech acts.
2. Much of the research has focused on a restricted set of illocutionary acts requests, apologies, and refusals In addition, complaints have been
investigated by Olshtain and Weinbach (1985) and Bonikowska (1988),
thanking by Eisenstein and Bodman (1986), invitations by Scarcella (1979),
suggestions by Bardovi-Harlig and Hartford (1990, 1993) and Rintell (1981),
compliments by Wolfson (1989b) and Billmyer (1990), greetings by DuFon
(2000), arguing by Adger (1987), complaints/criticisms by Murphy and Neu
(1996), and disagreements (Bardovi-Harlig and Salsbury 2004). Many of these
acts have two points in common. First, they constitute ‘relatively well-defined’
acts (De Beaugrande and Dressler 1981: 117) in the sense that they are
realized by means of a small set of easily recognizable linguistic elements
(many formulaic). For example, Wolfson (1983) has shown that nine syntactic
patterns account for 95 per cent of the compliments in her native-speaker
corpus. Second, many of these acts are face-threatening in nature and,
therefore, provide a means of studying to what extent L2 learners with
different L1 backgrounds are able to use native-like politeness strategies.
3
Task
Below are some examples of scenarios designed to elicit compliment responses from
Felasi (2007) together with an account of the differences in the responses of the native
speakers and non-native speakers.
1. In what ways do the non-native speakers’ responses reflect a) sociopramatic
failure and b) sociopgramalinguistic failure.
2. To what extent do you feel it is appropriate to define ‘pragmatic failure’ in
terms of differences between native and non-native speakers’ responses?
Scenario 1:
You have just finished presenting your research paper. At the end of the class (when you
were just leaving the classroom), one of your classmates say: “You did an excellent job! I
really enjoyed your presentation”. You answer: _____
Almost all of the NSs responses to this scenario were agreement. Their responses varied
between appreciation token “oh, thank you!”, to comment acceptance “thanks! I’m glad you
enjoyed it”, and praise upgrade “you have no idea how hard I worked for that!”. Only one NS
responded by disagreeing and questioning “Really? I thought it was just ok”. On the other
hand, almost all of the native speakers of Arabic (NSAs), have responded by either transfer
(returning the compliment) “oh, your presentation was much better”, or interpreting it as a
request “do you want me to help you with your presentation?” Only one of the NSAs
responded with a simple “thank you”. The responses of NNSs who took the English version of
the DCT showed more use of the agreement responses like “thanks!” and “It’s nice of you to
say so”. However, they also showed literal translation of Arabic formulaic expressions used as
compliment responses. These translations included “I’m your pupil”, which is a scale down
expression that means the speaker is much better than the addressee, and “I’m ashamed”
which might strike a NS as extremely out of place, but is literally translated from the widely
used Arabic formulaic expression (‫ ) أخجلتم تواضعنا‬akhjaltom tawaado’na.
Scenario 2:
You have some friends and relatives over for coffee and cake that you baked. Someone says:
“Tastes Yummy!”. You answer:_____
Eight of the ten NSs responded with “thanks” and then offering to give the speaker the
recipe “would you like the recipe?”. The other two responded by giving information or history,
“it’s a family recipe”. NNSs who answered in Arabic used questions “really? Did you really like
it?”, disagreement “no it’s not, you’re just complimenting me!”, and reassignment “my mom
gave me the recipe”. NSAs who answered in English did not use those two strategies in their
responses. Instead, they resolved again to literally translating Arabic formulaic expressions
like “Your taste is yummy”, and “I added my magic to it / that’s because I dipped my sweet
finger in it”. Only one person responded with “thank you”.
Scenario 3:
You were shopping for a skirt and a stranger (male) approaches you and says: “This would
look amazing on you!” You answer:____
NSs all responded with either no acknowledgment, or by not accepting the compliment
“what’s it got to do with you?”. NNSs who answered in Arabic and English also responded
with no acknowledgment, or by offending the man.
Scenario 4:
You were shopping for a skirt and a stranger (female) approaches you and says: “This would
look amazing on you!” You answer:___
4
NSs responded with either a question “really?/ oh, so do you think I should buy it?” or an
appreciation token “thanks / thank you”. NNSs who responded in Arabic used formulaic
expressions like “May Allah bliss you” jazaaki allah khair, “May Allah make all your days
beautiful” Allah yhalli ayyamek. NNSs who answered in English used tokens of appreciation
“thanks/ how sweet of you to say so”, or returned the compliment by translating Arabic
formulaic expressions “Your eyes are beautiful” oyoonech el helwa.
Scenario 5:
Some friends are over at your house. One of them looks at a clock hanging on the wall and
says: “I love your clock. It looks great in your living room!”. You answer: _____
NSs responded with comment history “It was a present from my daughter/ I bought it in
Harrods”, or acceptance “yes, I loved it when I bought it”. NNSs who answered in Arabic
interpreted this compliment as a request and responded with offering the clock to the speaker
and insisting that they take it. NNSs answering in English also insisted that the speaker take
the watch “you must take it! I swear, you must!”. Only one NNS answered with an
appreciation token “you like it! Thank you!”.
Scenario 6:
You’re wearing a new shirt and a colleague looks at you and says: “This shirt looks great on
you! Blue is a great color for you.” You answer: _____
NSs responded with appreciation tokens “thanks/ you made my day!”, questions “is it
really?/ do you think so?”, disagreement “I dunno, I prefer pink”, and comment acceptance
“oh, it’s my favorite color. Thanks”. NNSs who answered in Arabic responded with returning
the compliment “you’re more beautiful/ this is because you have a good taste”, disagreement
“thanks, but I know this is only a compliment”, and questions “really? Swear!!”. NNSs who
answered in English returned the compliment by translating Arabic formulaic expressions
“your eyes are beautiful and they see everything beautiful”, disagreement “please don’t say
that, you’re embarrassing me!”, scale down “oh, it’s so cheap! I bought it in the sales”, and
comment acceptance “thanks, I like it too”.
Felasi, H. 2007. Just Say “Thank You”: A Study of Compliment Responses. The Linguistics
Journal 2 (1):
5