Excerps from Anna Trosbor. 1997. Text Typology: Register, Genre and Text Type. Text Typology and Translation: 3-23. John Benjamins

Text typology

register, genre and text type

Which categories can be used to classify and explain ways in which types of discourse may be accounted for? Terminological problems concerning the distinction of text, discourse, register, genre, text type, discourse purpose, communicative purpose, rhetorical purpose and communicative function will be dealt with. A framework comprising a classifiction into registers and genres, with communicative function and text type as crucial categories within a discourse framework of field, tenor and mode will be suggested.

Text types

Virtanen (1990) has studied the difference in terminology use:

  • Grimes 1975
  • Sinclair & Coulthard 1975
  • Longacre 1982
  • Halliday & Hasan 1976
  • Quirk et al. 1985
  • Biber 1989

Trosborg (1997) proposes to use them interchangeably.

Text types cutting across registers and genres

Genre distinctions do not adequately represent the underlying text functions of English. Genres and texts types must be distinguished.

  • Texts within particular genres can differ greately in their linguistic characteristics (texts in newspaper articles can range from narrative and colloquial to informational and elaborated).
  • On the other hand, different genres can be similar linguistically (newspaper and magazine articles).
  • Linguistically distinct texts within a genre may represent different text types,
  • while linguistically similar texts from different genres may represent a single text type (Biber 1989:6).

While genres form an open-ended set (Schauber and Spolsy 1986), text types constitute a closed set with only a limited number of categories (also Chafe 1982, who proposes a four-way classification of texts, 'involvement-detachment' and 'integration-fragmentation').

Kinneavy (1971, 1980) classifies texts in terms of modes of how reality can be viewed. His text types are cognitive categories offering ways of conceptualizing, perceiving and protraying the world.

  • narration: our dynamic view of reality looks at change
  • evaluation: our dynamic view focuses at the potential of reality to be different
  • description: our static view focuses on individual existence
  • classification: focuses on groups

Based on cognitive properties, Werlich (1976) includes five idealized text types or modes (adopted by Hatim and Mason 1990, Albrecht 1995, Biber 1989 -based on linguistic criteria):

  • description: differentiation and interrelation of perceptions in space
  • narration: differentation and interrelation of perceptions in time
  • exposition: comprehension of general concepts through differentation by analysis or synthesis
  • argumentation: evaluation of relations between concepts through the extraction of similarities, contrasts, and transformations
  • instruction: planning of future behaviour
    • with option (advertisments, manuals, recepes)
    • without option (legislation, contracts)

The relationship between text types and genres is not straightforward.

Biber captures the salient linguistic differences among texts in English (see also Longacre 1976, 1982, and Smith 1985).

For 2,400 years there have been two traditions of classifying texts:

deriving from Aristotle's Rhetoric

Rhetoric often refers to the uses of language.
More specific, it refers to modes of discourse realized through text types
(narration, description, exposition, argumentation, etc.)
i.e. the classification of texts by type (Kinneavy 1980:3-4)

For others, it refers to communicative funcions as rhetorical strategies (Trimble 1985)

1) Classification according to purpose

In terms of communicative functions, is the discourse intended to

  • inform
  • express an attitude
  • persuade
  • create a debate

Whereas genre refers to completed texts, communicative functions and text types, being properties of a text, cut across genres.

  • informative texts:
    • newspaper reports
    • TV news
    • textbooks
  • argumentative texts:
    • debates
    • political speeches
    • newspaper articles
2) Classification according to type or mode
  • descriptive
  • narrative
  • expository
  • argumentative
  • instrumental

(Kinneavy 1980, Faigley & Meyer 1983)

The focus is on functional categories or rhetorical strategies, which is not normative but abstract knowledge

Longacre (1976, 1982), Smith (1985) and Biber (1989) refer to text types as "underlying shared communicative functions". Trosborg reserves these functions to a classification of speech acts according to a typology by Kinneavy, restricting text types to modes of discourse.

Other authors (but not Trosborg) take function as

  • "the kind of reality referred to" (Cassirer 1944, Urban 1939)
  • the level of social formality of a given discourse (Kenyon 1952)
  • nonmorphological classes of words in grammar (Fries 1952)

While communicative purpose is the aim of a text, rhetorical purpose is made up of strategies which constitute the mode of discourse realized through text types.

Text types are "a conceptual framework which enables us to classify texts in terms of communicative intentions serving an overall rhetorical purpose" (Hatim and Mason 1990:140).

Communicative functions

Purpose of discourse may depend on four factors of the linguistic process:

  • speaker
  • listener
  • message (thing referred)
  • lingustic material (text)

Aristotle proposed "a language concerned with things", and "a language directed to the hearer". A three-dimensional model of communication (triangle) was proposed by Bühler (1933).

Based on Aristotle and Bühler, a text can be classified into a particular type according to which component in the communication process receives the primary focus (Jakobson 1960, Kinneavy 1971).

  • speaker: expressive
  • listener: persuasive
  • world (thing referred): referential
  • language (linguistic code): literary

Roman Jakobson (1960) added two other uses:

  • metalanguage
  • phatic communication (to keep the channel open)

Jakobson's model was adopted by Dell Hymes (1974).


Kinneavy (1980:65) acknowledges Aristotle and Aquinas, Cassier, Morris, Miller, Russell, Reichenbach, Richards, Bühler, and Jacobson.

Reiss (1976) makes a typology of texts based on communicative functions for translation, similar to Nord (1997).


Speech acts

Speech act theory views language as action made up of a communicative act (Austin 1962, Searle 1969, 1976). Searle (1976) distinguishes six major classes of speech act:

  • assertives, eg. the speaker states the door is open and believes that it is.
  • directives, eg. the speaker gives the comman open the door and he wants the door to be opened.
  • commissives, eg. the speaker says I will open the door and intends to do it.
  • expressives (and evaluatives), eg. the speaker exclaims I like your coat and he means it.
  • declarations, eg. in saying I resign ore You're fired the speaker must have the role of employee or boss respectively. Also, the state described by the propositional content is realized by the very act of saying it. There is no psychological state needed.
  • representatives, eg. the judge declaring, I find you guilty as charged. Again, the state described by the propositional content is realized by the very act of saying it, and in addition there is the sincerity condition that the speaker must believe the proposition expressed.

The distinction of these general classes is based on four dimensions:

  • Illocutionary point (assertive, directive, commissive...)
  • Direction of fit (word-to-world WWL, or world-to-word WLW)
  • Psychological state (believes B, wants W, intends I)
  • Propositional content.

Each class divides into a number of different speech acts (eg. depending on whether the speaker is begging, asking, ordering or threatening; the illocutionary point is the same - influencing the hearer -, but different illocutionary forces are expressed). Acknowledging Traugott and Pratt (1980), Hatim and Mason (1990) adopted this framework for translation.

An advertisement may be predominantly referential, consisting of expressive (informative) statements, but the aim is persuading the consumer to buy, i.e. they are directive. This is why speech acts interrelate with sequences and conform the notion of illocutionary (or communicative) structure of a text.

Austin (1962) declared that speakers do not simply produce sentences that are true or false, but rather perform speech actions such as requests, warnings, assertions, etc. Searle (1969) adopted Grice's (1957) recognition of intention to his effort to specify the necessary conditions on the performance of speech acts.

Text pragmatics studies how sequences of speech acts are evaluated on the basis of higher order expectations about the text, and how these sequences of coherent microtexts contribute to the global coherence of a larger text (Ferrara 1985).

Text act

Text act: the predominant illocutionay force of sequences of speech acts must be recognized (Hatim and Mason 1990, Horner 1975).

Context focus

No theory of modes of discourse is rigid in its categorization, multiple views of reality and multiple types (Kinneavy 1980). Pure narration, description, exposition and argumentation hardly occur. A particular genre may make use of several modes of presentation. Text type focus or contextual focus refers to text type at the macro level, the dominant function of a text type in a text (Morris 1946, Werlich 1976, HM 1990, Virtanen 1992 'discourse type').

Two-level typology of text types and communicative functions:

Text types employed in a particular text (or genre) need not agree with its contextual focus. An argumentative text-type focus may be realized through narration, instructions may take the form of description, etc.

There is interaction between communicative purspose and rhetorical purpose (text type), eg. to persuade it is possible to narrate, describe, argue.


Genres reflect differences in external format and situations of use, and are defined on the basis of systematic non-linguistic criteria.

  • guidebook
  • nursery rhyme
  • poem
  • business letter
  • newpaper article
  • radio play
  • advertisement

Registers are divided into genres refelcting the way social purposes are accomplished in and through them in settings in which they are used. Bathia (1993:17) points out a science research article is an instance of scientific language as is an extract from a chemistry lab report.

Academic language shows in
  • casual chats
  • lectures
  • conversations
  • class
  • email
  • memos
  • scholarly papers
  • books
Legal register (language of law)
  • legislative texts
  • contracts
  • deeds
  • wills
  • judge declaring the law
  • judge/counsel interchanges
  • counsel/witness interchanges
  • textbooks
  • lawyers communications

In the case of restricted registers there is a close connection between register and genre, eg. weather forecasts.

Genre is a macrolevel concept, a communicative act within a discourseive network:

... repertoires of typified social responses in recurrent situations -from greetings to thank yous to acceptance speeches and full-blown, written expositions of scientific investigations - genres are use to package speech and make it recognizable to the exigencies of the situation (Berkenkotter & Huckin (1995)
Swales (1990) analyses the development of the concept genre in the fields of
  • folklore studies
  • literature
  • linguistics
  • rhetoric
Aristotle: genres as classes of texts

"a distinctive type or category of literary composition" (Webster)

Today genre refers to a distinctive category of discoruse of any type, spoken or written, with or without literary aspirations

The stuty of texts as genres, "how texts are perceived, categorized and used by members of a community" (Swales 1990:42), has attracted little attention from Linguistics (eg. Frow 1980), until the Systemic school put hands on it.

Rhetorical scholars have given genre a more central place, recently focused on social constitution of nonliterary forms of writing and speaking. Ethnographers concern about which labels are used to type communications, in order to reveal elements of verbal communication which are sociolinguistically salient (Saville-Troike 1982). In the field of LSP there has been growing interest in the sociocultural functions of disciplinary genres, eg. legal and scientific communication:

Genres are not simply assembies of similar textual objects, but coded and keyed events set within social communicative process (Todorov 1976, Fowler 1982, Swales 1990). "A rhetorically sound definition of genre must be centred not on the substance or form of the discourse but on the action it is used to accomplish" (Miller1984:151)
Genres embrace each of the linguistically ralized activity types which comprise so much of our culture (Martin 1985:250) Genre is a system for accomplishing social purposes by verbal means. It "refers to the stages purposeful social processes through which a culture is realized in a language" (Martin and Rothery 1986:243)

Communicative purpose as the defining criterion of genre

For some scholars genres are defined on the basis

Swales emphasizes the socio-rhetorical context of genre, the categories are those of the community, and communicative purpose is the defining criterion. Genre as a social action operates as a mechanism to clarifying what communicative goals are.

A multi-dimensional approach to genre

Unclear relation between genre and register (Ventola 1984). Is genre a system underlying register? For Trosborg (1997) it is not.


In the narrow sense of occupational field, genres such as
  • contracts will be part of legal register
  • a sermon will involve the religious register


but, a particular genre may cut across a number of registers
  • a research article in chemistry may be similar
  • to a research article in sociology
(Swales 1981)
One register may be realized through various genres, in this sense genres are subordinated to registers. Conversely, one genre may be realized through a number of registers just as a genre constrains the ways in which register variables of field, tenor and mode can be combined.
Registers impose constraints at the linguistic level of vocabulary and syntax. Genre constraints operate at the level of discourse structure.

Genre specifies conditions for beginning, structuring and ending a text.

Genres can only be realized as completed texts (Couture 1986)

Trosborg (1997) sees genres as having complementary registers. Communicative success of a text may require appropriate combinations of genre and register (Couture 1986). In agreement with the stand taken by Swales (1990), Bhatia (1993) takes genre analysis form linguistic description to explanation: Why do members of a specialist community write the way they do? Berkenkotter & Huckin (1995) develop a sociocognitive theory of genre, which Trosborg (1997) applies as an explanatory approach to hybrid political texts from the EU.

Genres cannot be identified by communicative purpose, eg. poetic genres aimed at giving verbal pleasure defy ascripton of communicative purpose. Medium of communication may also be decisive: memos, emails, faxes.

Model: texts form part of communicative situations. Hallidays (1971) functional approach with three-fold division (used by Vermeer, Nord, Hatim & Mason 1990, and Baker 1992 for translation, and by Bhatia 1993 for SPL).

A genre can only be accounted through a specification of field, tenor and mode and a description of the linguistic features realized in the ideational, inerpersonal and textual components of particular texts (Eggins 1994). Kussmaul 1997 shows how a change of a single parameter may result in a change of genre.


Varieties of language use have been referred to as registers (Reid 1956, Halliday et al. 1964)

Halliday et al. 1964 divided language into

user-related varieties or dialects (Corder 1973)
  • geographical
  • temporal
  • social non-standard dialects
  • idiolects
use-related varieties or registers of occupational fields
  • religion
  • legal documents
  • newspaper reporting
  • medicine
  • technical reporting

These are some approaches to the study of register:

Trosborg (1997) concludes that register is too broad a notion. Focusing on the language of a field, register analysis disregards differences of genres within the field. Labels such as legal English are misleading and overprivilege a homogeneity of content at the expense of variation in communicative purpose, addresser-addresee relationships and genre conventions (Swales 1990:3).

In order to understand how texts organize informationally, rhetorically and stylistically, there is the need of other considerations on top of pure textual evidence.

Inappropriate usages of the term register:

Main references

Vijay K. Bhatia. 1993. Analysing Genre. Language use in professional settings. Longman.

Douglas Biber. 1989. A Typology of English Texts. Linguistics 27: 3-43.

James L. Kinneavy. 1980. A Theory of Discourse. Norton.

M.A. K. Halliday & R. Hasan. 1976. Cohesion in English. Longman

Basil Hatim & Ian Mason. 1990. Discourse and the Translator. Longman.

John M. Swales. 1990. Genre Analysis. English in academic and research settings. Cambridge University Press.

John M. Swales. 2000. Further Reflections on Genre and ESL Academic Writing. Symposium on Second Language Writing. West Lafayette, Indiana, USA September 15-16, 2000

Although many would likely concur with Bakhtin's dictum that "The better our command of genres, the more freely we employ them," operationally genre remains a disputed framework for ESL writing courses and approaches. Controversies polarize around repression versus expression, individual voice versus conventionalized pattern, imitative play versus contextual realpolitik, specific guidelines versus general principles, and cultural subordination versus cultural resistance. Recent work confirms the contested nature of the theoretical ground. On the one hand, Johns (2000) offers several example of genre-based approaches in effective action; on the other hand, Freedman (2000) questions whether EAP instructors can sufficiently escape their own classroom contexts to offer real assistance with the genres of the wider academy. In this presentation, I discuss these controversies through the lens of new advanced materials for NNS graduate students (Swales & Feak, 2000) premised on cross-disciplinary "difference," participant disciplinary analysis, genre systems, and a task taxononmy privileging rhetorical reflection. I argue that while border crossings may be hazardous with undergraduate "school genres," and certainly in preparing students for writing at work (Freedman, 1993), they are less so in research genres. Reasons for this include the public nature of many research genres, the established evaluative processes that adjudicate them, and student capacity to assess the appropriacy of any advice offered.