Discourse Completion Task

A discourse completion task (DCT) is a written data elicitation instrument in which participants read a description of a social situation and write what they would say in that context. It is the standard tool for studying speech act production and pragmatic competence across languages and cultures.

Origin

Shoshana Blum-Kulka and colleagues developed the DCT for the Cross-Cultural Speech Act Realization Project (CCSARP) in the 1980s. The CCSARP used DCTs to compare how speakers of different languages performed requests and apologies, establishing the instrument as the default method in interlanguage and cross-cultural pragmatics research.

Format

A typical DCT item consists of:

1. Situation description: specifying the setting, interlocutor relationship, and social variables (power, distance, imposition)
2. Prompt: a dialogue turn preceding the target speech act
3. Blank line: where the participant writes their response
4. Optional rejoinder: a follow-up turn after the blank, constraining the type of response

Example:

You need to ask your professor to extend a deadline for an assignment. You arrive at the professor's office hours.

Professor: "What can I help you with?"

You: _______________

What DCTs Reveal

• Speech act strategies (e.g., direct vs indirect requests)
• Pragmatic transfer from L1 (Language Transfer)
• Sensitivity to social variables (power, distance, degree of imposition)
• Developmental changes in pragmatic competence across proficiency levels
• Cross-cultural variation in speech act realisation

Strengths

• Efficiency: large amounts of comparable data collected quickly
• Control: social variables (power, distance) can be systematically manipulated
• Comparability: the same situations administered to speakers of different languages
• Coding ease: written responses are readily analysable

Limitations

• Ecological validity: what people write they would say may not match what they actually say in spontaneous interaction (see Ecological Validity)
• Lack of interaction: real speech acts unfold over multiple turns with negotiation; DCTs capture only a single turn
• Written modality: responses may be more formal, planned, and edited than spoken language
• Stereotype effect: participants may produce what they think is socially appropriate rather than what they would genuinely say
• Missing prosody and non-verbal cues: intonation, facial expression, and gesture are absent

Alternatives and Complements

• Oral DCTs: participants respond orally, adding prosodic information
• Role plays: interactive, multi-turn data (higher ecological validity but harder to control)
• Natural observation: recording authentic speech acts in context (highest validity, lowest control)
• Multimedia elicitation: video-based scenarios that provide richer context

References

• Blum-Kulka, House & Kasper (1989): Cross-Cultural Pragmatics: Requests and Apologies (CCSARP)
• Kasper & Dahl (1991): comparison of data collection methods in pragmatics
• Golato (2003): comparing DCT and naturally occurring compliment responses
• Bardovi-Harlig (2010): pragmatics research methods review