Vanity Fea

Aspectos metadiscursivos en el discurso académico

jueves 13 de mayo de 2010

Aspectos metadiscursivos en el discurso académico

Notas sobre una conferencia de Annelie Ädel, de la Universidad de Estocolmo, invitada hoy en nuestro departamento. Puede encontrarse un artículo completo sobre este trabajo de investigación en el último número del Nordic Journal of English Studies, un volumen sobre metadiscurso editado por ella y por Anna Mauranen.

Pongo al final mis notas en inglés. La charla se titulaba, en inglés y sobre el inglés—

Just to give you kind of a map of where we are going: Metadiscourse in Spoken and Written Academic English".

This is a contribution to the study of linguistic reflexivity and it will follow these steps:

a) a discussion of the present model
b) a sketch of previous reserch in academic metadiscourse
c) the material (corpora) used
d) the method of study
e) a proposed taxonomy of metadiscourse types and functions
f) a comparison of spoken and written metadiscourse
g) a conclusion

a) A discussion of the present model

Metadiscourse is discourse about discourse, language used to talk about talk. It may refer to communicators (Speaker and Hearer, interactants), or to the situation in which communication takes place. Lyons defined linguistic reflexivity as "the capacity of natural language to refer to itself" (1977: 5). Hockett argues it is a language universal (1977: 173) and Silverstein notes that it is unique to human communication systems (1976: 16). Verschueren (1999: 187-188) considers metadiscourse a fundamental dimension of discourse. The inaugural classical approach is Jakobson's inclusion of the metalinguistic function among his scheme of the several functions of language.

In applied linguistics, metadiscourse has been the object of study since late 1980s (Vande Kopple, Crismore, Markkanen et al., Mauranen, Hyland, Ädel)... It is nonetheless a divided area: problems of definition are a major issue. But 2 strands in definition: interactive vs. reflexive models of metadiscourse.

The interactive model
It uses a very broad definition of metadiscourse, with a bias towards written communication. Centered on the study of authorial stance, intertextuality, all types of connectives… For Ädel, it tries to subsume too many features or categories under the umbrella of metadiscourse.

The reflexive model
It uses a narrower definition, sticking rather more closely to the meta- notion: based on reflexivity. Restricted to elements of discourse which are about the ongoing discourse. Here we'll stick to this (perhaps biased) view of metadiscourse as reflexivity.

Anyway we use here fuzzy categories: instances analyzed are often multifunctional, and there is no absolute borderline between what is and what is not metadiscourse.

Ädel's definition (2006): metadiscourse is discourse "referring to the evolving discourse itself or its linguistic form". References should be to the world of discourse, not the "real world" in which the discourse activity takes place. Reference to the speaker-writer should be (in order to count as metadiscourse) to that speaker qua speaker-writer, audience qua audience—not in other possible capacities or social roles which may be superimposed.

Personas in the world of discourse: It is useful to think about metadiscursive reference in terms of personas. The speaker-writer can appear in many personas, only some of which are metadiscursive: e.g. he can also be simultaneously the organiser of the discourse activity, or an experiencer in the real world, a US citizen, a fellow human being... but these are different personas or capacities.

The use of "We" in lectures provides some examples. One may refer using "we" to communicators in ongoing discourse, or to other kind of personae ("we" as citizens, or as members of the human species..). But only those related to the ongoing interactive communication are metadiscursive uses of "we".

b) Previous research.

As regards academic discourse, research on metalanguage has been carried out almost exclusively on written language. There is very little done on spoken and written metadiscourse simultaneously. Luukka, Mauranen, Pérez-Llantada, Thompson, address spoken genres (conferences, lectures). And Mauranen & Luukka deal with both speech and writing.

Mauranen (2001) differentiates monologic, dialogic, or interactive types of metadiscourse (depending on whose talk is being commented on). She splits spoken and written discourse, arguing that they are different modes.

Luukka (1994) mentions the following functions: Textual (relative to the structuring of text), interpersonal (signaling of attitudes of participants), contextual… This is an interactive model, attentive to the speakers' stance. Lukka provides a "lumping" approach, dealing simultaneously with spoken and written discourse, but this choice is not commented on. Comparisons are provided between the oral/written versions of the same talk.

Ädel puts forward a proposal toward a new taxonomy—which should be attentive to specific discourse functions, and should be based on speech and writing (a "lumping" approach).

"Discourse function essentially refers to the rhetorical function that the metadiscursive expression performs in its immediate discourse context" (Ädel 2006: 57ff).

c) The material for study

The corpus for spoken discourse in this study consists of 30 large university lectures from the MICASE corpus, 255.000 words in all. For written discourse, 130 papers from MICUSP, a corpus of writing by senior university students and graduate students. 400,000 words, covering several disciplines.

d) the method:

Corpus linguistics methods were used to retrieve potential examples of metadiscourse, eg. I, we, you,—followed by a manual sifting of irrelevant (non-metadiscursive) instances. Other disregarded data include those coming from quotes or arising from speech dysfluencies, hesitations, etc.

Personal metadiscourse includes reference to discourse participants. Impersonal metadiscourse does not involve any explicit reference to participants. Here, only PERSONAL types of metadiscourse were dealt with.

The following taxonomy was based on 5,000 examples for spoken, 800 examples for written data. (But only one analyst...).

e) A taxonomy of discourse functions:

—a provisional proposal. Two big subtypes: functions relating to metatext, and functions relating to audience interaction.

1) Metatext (Metalinguistic comments, Discourse organisation, Speech act labels).

2) Audience interaction (references to the audience)

Metatext is oriented towards code, language, discourse… Audience interaction functions refer to the audience.


Metatextual functions can be classified under three headings: metalinguistic comments, discourse organisation, and speech act labels.

1.1. Metalinguistic comments

1.1.1. Repairing:

Corrections to discourse, both self- or other-initiated. (There are NO examples of discourse repairing in written data).

1.1.2. Reformulating:

This includes the offering of alternative terms or expressions, because of their added value of expanding previous discourse.

1.1.3. Commenting on linguistic form/meaning (choice of words, meaning…)

1.1.4. Clarifying (spelling out the addresser's intentions to avoid misinterpretation)

1.1.5. Managing terminology (giving definitions, providing terms or labels for phenomena talked about).

(Examples of all these were given in the talk and can be found in the paper).

1.2. Discourse organisation

Ten functions are distinguished. The first 5 manage the topic; 6-10 manage the phorics.

1.2.1. Introducing a topic.

1.2.2. Delimiting a topic (explicitly stating in which way the topic is constrained in the discussion. (see e.g. d) above).

1.2.3. Adding to a topic (Used to explicitly comment on the addition of a sub-topic).

1.2.4. Concluding a topic (closing the topic), e.g. "We've talked in detail…", "I have attempted in this paper to…" etc.

1.2.5. Making asides—opening or closing a 'topic sidetrack' or digression. (No examples of this function were found in the written data).

1.2.6. Enumerating: shows how specific parts of the discourse are ordered in relation to each other.

1.2.7. Endophoric marking: points to a speciific location in the discourse, refers to cses in which it is not clear whether what is referred to occurs before or after the current point – e.g. instructions to look at tables, or handouts…)

1.2.8. Previewing: points forward in discourse

1.2.9. Reviewing: points backwards, reminding what has already taken place in the discourse.

1.2.10. Contextualising: Comments on the conditions of the situation of writing or speaking, and thus containing traces of the production of the discourse (typically, spelled-out justifications for choices made in planning or organising the discourse). (Also found in written discourse: comments on layout or printing, on ways to navigate hypertextual links, etc.).

1.3. Speech act labels:

1.3.1. Arguing (for or against an issue)

1.3.2. Exemplifying (used when explicitly introducing an example).

1.3.3. Other speech act labels….
Other speech acts were found not to be frequent or common enough to give them separate categories. But the subtypes could be extended in specific genres or modes.

2. AUDIENCE INTERACTION (references to the audience)

2.1. Managing comprehension (channel) – No examples of this function are found in the written data.

2.2. Managing audience discipline (cases in which audience is directly addressed an instructed to do something, reprimanded or complimented). – No examples are found in the written data either.

2.3. Anticipating the audience's response (e..g. attributing statements to the audience as potential objections or counterarguments).

2.4. Managing the message:
Emphasises the core message in what is being conveyed, provides big picture or main poiint, what the audience should remember or experience based on the discourse. Also when addresser refers explicitly to the desired uptake.

2.5. Imagining scenarios:

A mutual thought experiment, asking the audience to see things from a specific perspectiv, opening up another world in the world of discourse.

(Examples were likewise provided for all of these categories…)

f) A comparison between metatextual functions in spoken and written academic discourse

A number of similarities as well as differences between spoken/written discourse were found. 23 functions were identified, most found in both sets of data. But there are salient data as regards their distribution.

7 discrepancies between speech and writing
5 of those related to different conditions in S/W
2 have to do with genre differences.

Discrepancies: can be attributed to the differences between prototypical speech and prototypical writing.

Repairing, marking asides, contextualising, managing channel and comprehension, managing audience discipline... are typical of spoken discourse and are missing in written data.

The presence of repairing or marking asides in spoken discourse is attributable to lack of time in real-time discourse—and is missing in writing.

Contextualising is marginal in writing, much more common in spoken lecturers. Marks of ad-libbing in talking, improvising discourse. E.g. references to remaining time, etc.: Temporal constraints are quite often commented on in lectures.

Managing comprehension, channel and audience discipline are possible in spoken discourse, and frequent; they could occur in writing, but they are rarer.

The importance of time and interaction. Both dimensions are important constraints in any linguistic production (Chafe 1982): the time available, and the modes of interaction available— 2 constraints marking all discourse. Regarding these, there are two constraints in speech which do not apply to writing: lack of time for revision, and the immediate presence of an audience.

Discrepancies in genre: (2) —arguing, and managing the message. These might be conditioned by the nature of the data used. Genres used have different purposes and involve different speaker-writer roles. Arguing is more common in written language, and the differences in data are also likely to be genre-related.

Managing the message is quite common in lectures, an oral genre, but rare in writing. Here too we find genre-related factors involving power relations (teacher-students).

g) Conclusions

A lumping approach makes possible a comparison between spoken and written discourse, showing their differences but also what they have in common—which is a lot.

Spoken metadiscourse appears to perform a greater range of discourse functions than written metadiscourse. Or perhaps the speakers (lecturers) happen to have a more advanced rhetorical repertoire. More studies are needed.

It would be interesting to study these functions in conversational interaction. Future research lines: Statistical differences would need to be studied, categorizing relative frequencies of functions in different modes. Likewise, more specific studies across modes and across speaker groups (teachers vs students, professional writers vs beginners…) would be welcome.

Also, only personal types of metadiscourse have been addressed here. Impersonal types may differ in speech and writing; it is to be expected that spoken discourse will be more personal.

Research should focus on specific forms used to carry out these functions (here the focus has fallen on functions). Many specific expressions are likely to play a role as stored units which help people process discourse, develop expectations, etc.


Ädel, Annelie, and Anna Mauranen, eds. Nordic Journal of English Studies 9.2 (2010). Special issue on metadiscourse. (Introd.: "Metadiscourse: Diversity and Division").
Ädel, Annelie, and Ute Römer. "Research on Proficient Student Writing across Disciplines and Levels: Introducing MICUSP." Paper given at ICAME 30, Lancaster, UK, May 2009.
Ädel, Annelie. Metadiscourse in L1 and L2 English. Amsterdam and Philadelphia: John Benjamins, 2006.
Baron, Naomi. "The Written Turn." English Language and Linguistics 9.2 (2005): 359-76.
Biber, Douglas. Variation across Speech and Writing. Cambridge: Cambridge UP, 1988.
Chafe, Wallace, and Deborah Tannen. "The Relation between Written and Spoken Language." Annual Review of Anthropology 16 (1987): 383-407.
Chafe, Wallace. "Integration and Involvement in Speaking, Writing, and Oral Literature." Spoken and Written Language: Exploring Orality and Literacy. Ed. Deborah Tannen. Norwood (NJ): Ablex, 1982. 35-53.
Chaudron, Craig, and Jack C. Richards. "The Effect of Discourse Markers on the Comprehension of Lectures." Applied Linguistics 7 (1986): 113-27.
Crismore, Avon. Talking with Readers: Metadiscourse as Rhetorical Act. New York: Peter Lang, 1989.
Goldberg, Adele E., and Devin Casenhiser. "English Construction." The Handbook of English Linguistics.
Hockett, Charles. The View from Language: Selected Essays 1948-1974. Athens (GA): U of Georgia P, 1977.
Hyland, Ken. Metadiscourse: Exploring Interaction in Writing. London and New York: Continuum, 2005.
_____. "Persuasion and Context: The Pragmatics of Academic Discourse." Journal of Pragmatics 30 (1998): 437-55.
Jakobson, Roman. On Language: Roman Jakobson. Ed. L. R. Waugh and M. Monville-Burstion. Cambridge (MA) and London: Harvard UP, 1998.
_____. The Framework of Language. Michigan: Michigan Studies in the Humanities, 1980.
Luukka, Minna-Riitta. "Metadiscourse in Academic Texts." In Text and Talk in Professional Contexts: Selected Papers from the International Conference "Discourse and the Professions," Uppsala, 26-29 August, 1992. Ed. B.-L. Gunnarsson, P. Linell, and B. Nordberg. Uppsala: ASLA, The Swedish Association of Applied Linguistics, 1994. 77-88.
Lyons, John. Semantics. Cambridge: Cambridge UP, 1977.
Markkanen, Raija, Margaret Steffensen and Avon Crismore. "Quantitative Contrastive Study of Metadiscourse: Problem in Design and Analysis of Data." Papers and Studies in Contrastive Linguistics 28 (1993): 137-52.
Mauranen, Anna. "Reflexive Academic Talk: Observations from MICASE." Corpus Linguistics in North America: Selections from the 1999 Symposium. Ed. R. Simpson and J. Swales. Ann Arbor (MI): U of michigan P, 2001. 165-78.
Miller, Jim. "Spoken and Written English." The Handbook of English Linguistics. Ed. B. Aarts, and A. McMahon. Oxford: Blackwell, 2006. 670-91.
Pérez-Llantad Auría, Carmen. "Signalling Speaker's Intentions: Toward a Phraseology of Textual Metadiscourse in Academic Lecturing." In English as a Glocalisation Phenomenon: Observations from a Linguistic Microcosm. Ed. C. Pérez-Llantada and G. R. Ferguson. Valencia: U of Valencia, 2006. 59-88.
Roberts, Celia, and Brian Street. "Spoken and Written Language." Blackwell Handbook in Linguistics. Ed. F. Coulmas. Oxford and Malden: Blackwell, 1998. 168-86.
Silverstein, Michael. "Shifters, Linguistic Categories, and Cultural Description." In Meaning in Anthropology.
Simpson, Rita, Sarah Briggs, Janine Ovens, and John Swales. The Michigan Corpus of Academic Spoken English. Ann Arbor (MI): U of Michigan, 1999. (MICASE).
Thompson, Susan Elizabeth. "Text-Structuring Metadiscourse, Intonation and the Signalling of Organisation in Academic Lectures." Journal of English for Academic Purposes 2 (2003): 5-20.
Vande Kopple, William J. "Metadiscourse and the Recall of Modality Markers." Visible Language 22 (1988): 233-72.
_____. "Some Exploratory Discourse on Metadiscourse." College Composition and Communication 26 (1985): 82-93.
Verschueren, Jef. Understanding Pragmatics. London: Arnold, 1999.
Ed. B. Aarts and A. McMahon. Malden (MA): Wiley / Blackwell, 2006. 343-55. Ed. K. Basso and H. Selby. Albuquerque: U of New Mexico P, 1976.

Algunos elementos metaficcionales

0 comentarios