A01 cohe4573 01 se fm. Qxd

4 3
F I G U R E 3 . 2
tude of the imposition in this case) may affect the way that the request is
crafted (see Figure 3.2, below).
Data elicitation, such as through DCTs and role-play, tends to be a quick
and convenient way to collect a relatively large amount of language data. 
At the same time, pragmatics researchers have pointed out repeatedly that
elicited data, particularly written tasks such as the written DCT, do not 
accurately reflect the way we actually speak.
11
For example, the speakers 
providing language data may not be familiar with the situations described in
the scenarios. Or even if they are, real-life consequences (consequentiality
12
)
are lacking in such imagined contexts, and we may be testing their acting
skills along with pragmatic ability. The number of studies that researched
effects of the DCT in fact have questioned the validity of DCT-elicited data
and have recommended caution in using these data.
13
Data collected from
DCTs may reflect the speakers’ knowledge of pragmatics rather than their
online productive skills.
14
Speakers providing pragmatic language use might
consider the social desirability of what should be said and how it should be
said, and demonstrate more of what they perceive as appropriate behavior
than what they would actually say in interactive contexts. Also, DCTs may
be a valid source of pragmalinguistic
15
(language-focused) data, but might
not be reflective of the sociopragmatic (culture-focused) aspects of the speech
act.
16
Some have argued that the effects of the DCT may differ depending on
11
Kasper and Dahl (1991); Kasper and Rose (2002). Also see Golato (2003); Kasper
(2000) for more discussion.
12
Bardovi-Harlig and Hartford (2005).
13
See for example, Beebe and Cummings (1996); Golato (2003); Hartford and
Bardovi-Harlig (1992); Kasper and Rose (2002).
14
Golato (2003).
15
Though somewhat overlapping, sociopragmatics deals with sociocultural norms 
for linguistic behavior in a given context while pragmalinguistics refers to the actual
language forms used to convey the intended meaning (Thomas, 1983).
16
Nelson et al. (2002).

---