The Multidimensional Approach

There are many studies that describe the

situational and linguistic characteristics of registers.
One of these studies is multidimensional approach.

The multidimensional(MD) approach to register

variation was developed by the American linguist
Douglas Biber (1988) to provide comprehensive
descriptions of the patterns of register variation in a
language.
 Biber and Conrad (2001:184) state that an MD
analysis includes two major components :

1) identification of the underlying linguistic

parameters, or dimensions of variation
2) specification of the linguistic similarities and
differences among registers with respect to those
dimensions
 The MD approach has three major distinguishing
characteristics :
1) the use of computer-based text corpora to
provide a broad representation of the registers in a
language
2) the use of computational tools to identify
linguistic features in texts
3) the use of multivariate statistical techniques to
analyze the co-occurrence relations among
linguistic features, thereby identifying underlying
dimensions of variation in a language.
 Register studies and the MD approach have also
underlined the fact that the distribution of
grammatical structures is different across various
text types. The reason for this variation among
different text types is, as Lemke explains, “where
the field of activity differs, there are characteristic
(and also statistically consistent) differences in the
frequencies of grammatical patterns which in turn
reflect differences in communicative purposes”
(1995:27).
 As Biber and Conrad also state (2001 : 176),
functional descriptions based on texts without
regard for register variation are inadequate and
often misleading : for register descriptions, a
comparative register perspective provides the
baseline needed to understand the linguistic
characteristics of any individual register.
 The first example of the MD approach is Biber’s (1988)
own study. In this study, various spoken and written
registers in English are compared along dimensions of
linguistic variation. The following six dimensions are
employed to make comparisons among registers :
1)involved discourse versus informational discourse
2)narrative concerns versus non-narrative concerns
3)situation dependent reference versus explicit
reference
4)overt expression of persuasion
5)abstract discourse versus non-abstract discourse
6)on-line information production
 Various other register studies in English were carried
out, using the MD approach. For example, Atkinson (1992)
studied the historical evolution of medical research writing
in terms of four dimensions proposed by Biber.
Conrad (1996a ; 1996b) applied the MD modal of
variation in English to compare professional research
articles, university-level textbooks, and university student
papers in biology and history.
Reppen (1994 ; 1995) used MD approach to study
spoken and written registers used by elementary school
students in English.
 The MD approach has also been used to investigate the
patterns of register variation in nonwestern languages.For
instance, Biber and Hared (1994) investigated register
variation in Somali using three dimensions.
Kim and Biber (1994) studied register variation in
Korean along with six dimensions.
Kessapidu (1997) analyzed the persuasion patterns of
Greek business letters in terms of five dimensions of the
MD approach.
 In Turkish, the studies that applied the MD
approach are few. Bayyurt (2000) used the MD
approach to compare various spoken and written
registers in terms of formality.
Özyıldırım (2010) applied five dimensions of the
MD approach to study the discoursal features of the
six different registers in Turkish.
The MD approach has also been used to
investigate the patterns of Turkish official language
by Yarar (2002).
 Both theoretical assumptions and major components of
the MD approach indicate that there are three key terms of
the approach as linguistic co-occurrence, dimension and
multiple dimensions. The first of these terms, linguistic co-
occurrence, is considered as central in MD approach since a
register is characterized by a set of co-occurring linguistic
features.
Dimension, on the other hand, involves a group of
linguistic features, which co-occur with a markedly high
frequency in texts. Thus, dimension is used to analyze the
linguistic co-occurrence. However, not a single dimension
but multiple dimensions are employed in the MD approach.
 To give an example, Dimension 1 (involved/
interactional discourse versus informational
discourse) represent discourse with interactional,
affective, involved purposes versus discourse with
highly informational purposes, which is carefully
crafted and highly edited (Biber 1988:115).

Some of the lexico-grammatical features of

Dimension 1 are outlined in Özyıldırım (2010:41):
 Positive and negative features of Dimension 1
Interactional (unplanned) discourse
(positive features)
1st. Person pronouns
2nd. Person pronouns
Direct questions
Causative adverbial subordinators
Wh-complement clauses
Emphatics
Amplifiers
Discourse particles
Informative (planned) discourse
(negative features)
Nouns
Prepositions
Stative forms
Agentless passives
Relative Clauses
Adjectives
 There are two groups of features in Dimension 1,
labeled positive and negative.
The positive features represent discourse with
interactional, affective and involved purposes whereas
negative features represent discourse with highly
informational purposes, which is carefully crafted and
highly edited (Biber, 1988 :115).
Furthermore, the two groups have a complementary
relationship. That is, if a text has frequent occurrences of
the positive group of features, it will have markedly few
occurrences of the negative group, and vice versa.
 Positive features of Dimension 1 are mostly observed
with high frequencies in interactional, spoken and informal
discourse types such as interviews, telephone
conversations, spontaneous conversations, TV ads, etc. For
example, first person pronouns, second person pronouns,
direct questions require a specific addressee and indicate a
high degree of involvement with that addressee and thus
require interaction.
 Similarly, causative adverbial subordinators indicating
causes of things or actions; Wh-complement clauses
resembling questions, emphatics which mark the presence
of certainty towards a proposition by way of certain words
like ‘really’, ‘for sure’, ‘of course’ and amplifiers which are
used to indicate the reliability of propositions positively
such as ‘very’, ‘completely’, etc. and discourse particles like
‘well’ ‘anyway’, ‘anyhow’ all serve interactional functions
and are used frequently in spoken registers.
 On the other hand, negative features of Dimension 1
are mostly observed with high frequency in written and
formal registers such as academic articles, legal texts, news
reports, textbooks, etc.
For example, Biber(1988:227) states that, a high
frequency of nouns in a text indicates a high abstract and
informational focus, as opposed to primarily interpersonal
or narrative foci. Thus, high nominal content clearly shows
highly abstract, nominal, conceptual and informational
focus in a text.
 Prepositions, relative clauses, adjectives are also used
for expanding and elaborating the information presented in
a text and provide detailed, elaborated and technical
information.
Passives are taken as one of the important markers of
written language. In passive constructions, the agent is
demoted or dropped altogether, resulting in a static, more
abstract presentation of informationand a decontextualized
style.
 Stative forms or sentences with nonverbal predicates
might be considered as the markers of a static,
informational style common in writing since they preclude
the presence of an active verb (Biber, 1988 : 228).

As a result, all these negative features of Dimension 1 are

the indicators of written, formal and informational text
types.
 In addition to the descriptions of a single register, a
corpus-based approach enables comparative analyses of
register variation. One advantage of a comparative register
perspective is to understand the linguistic characteristics of
a particular register relative to a representative range of
registers in the language.
As Biber and Conrad (2001:179) state, most
grammatical features are distributed in very different ways
across registers. These overall distributional patterns
correspond to the differing production circumstances,
purposes and levels of formality found across registers.
 For example, in his book Variation across Speech
and Writing(1988:172) Biber compared several
registers in terms of the features of Dimension 1. In
this study, the most interactional register with the
highest positive feature score is found to be as face-
to-face conversations whereas the most
informational register with the highest negative
feature score is found to be official documents.
 Similarly, Özyıldırım (2010:184) in her study
compared six different Turkish registers in terms of
Dimension 1. In the corpus of this study, the most
informative text type is found to be as legislative
texts. It is followed by scientific research articles,
newspaper feature articles, man/woman magazines
and stand-up shows in this order. The most
interactional text type in this corpus is found to be
as TV commercials.
 As stated by Kim and Biber (1994, p.157), in
multidimensional analysis, both a microscopic and
macroscopic approach are used. Microscopic
approach focuses on the discourse functions of
individual linguistic features in particular registers
while a macroscopic approach seeks to define the
overall parameters of variation among registers.
 Microscopic and macroscopic analyses have
complementary strengths in that a microscopic analysis can
pinpoint the exact communicative functions of individual
linguistic features in particular registers, but it does not
provide the basis for overall generalizations concerning
differences among registers.
In contrast, the macroscopic analysis focuses on the
overall patterns of variation among registers, building on
previous microanalyses to interpret those patterns in
functional terms.
 Conclusion
Unlike most corpus-based research, MD studies
investigate language use in individual texts. This approach
describes how linguistic features co-occur in each text,
resulting in more general patterns of linguistic co-
occurrence that hold across all texts of a corpus. The
approach can thus be used to show how patterns of
linguistic features vary across individual texts, or across
registers and genres.
Finally, the MD approach will also provide a framework
for additional cross-linguistic investigations, eventually
allowing identification of universal tendencies.
 References
Atkinson, D. (1992). The Evolution of Medical Research Writing from 1735 to 1985: The Case Of the Edinburgh Medical
Journal. Applied Linguistics 12 (4), 337-374.
Bayyurt, Y. (2000). Türkçe’de Resmiyet Kavramına TV Sohbet Programları Çerçevesinden Bir Bakış. Dilbilim Araştırmaları
2000, 17-37.
Biber, Douglas, 1988. Variation Across Speech and Writing. Cambridge University Press, Cambridge.
Biber, Douglas, Conrad, Susan, 2001. Register Variation : A Corpus Approach. In : Deborah Schiffrin,Deborah Tannen and
Heidi Hamilton (Eds.), The Handbook of Discourse Analysis. Blackwell Publishers, Oxford, pp.175-196.
Biber, Douglas, Hared, Mohamed, 1994. Linguistic Correlates of the Transition to Literacy in Somali : Language Adaptation
in six Press Registers. In : D. Biber and E. Finegan (Eds.), Sociolinguistic Perspectives on Register. Oxford University
Press, Oxford, pp. 182-216.
Conrad, S. (1996a). Academic Discourse in two Disciplines : Professional Writing and Student Development in Biology
and History, Unpublished PhD. Dissertation, Northern Arizona University.
Conrad, S. (1996b). Investigating Academic Texts with Corpus-based Techniques : An Example from Biology. Linguistics
and Education 8, 299-326.
Kessapidu, S. (1997). A Critical Linguistic Approach to a Corpus of Business Letters in Greek. Discourse and Society 8,
479-500.
Kim, Y.J., Biber, D., 1994. A Corpus-based Analysis of Register Variation in Korean. In : D. Biber and E. Finegan (Eds.),
Sociolinguistic Perspectives on Register. Oxford University Press, Oxford, pp. 157-161.
Lemke, J. L. (1995). Textual Politics : Discourse and Social Dynamics, London : Taylor and Francis.
Özyıldırım, I. (2010). Tür Çözümlemesi: Türkçe Metin İncelemeleri ve Karşılaştırmalar. BilgeSu, Ankara
Reppen, R. (1994). Variation in Elementary Student Language : A Multidimensional Perspective, Unpublished Ph.D.
Dissertation, Northern Arizona University.
Reppen, R. (1995). A Multi-dimensional Comparison of Spoken and Written Registers Produced by and for Students. In
Brita, W., Tanskanen, S.K. and Hiltunen, R. (eds.), Organizations in Discourse (Proceedings from the Turku Conference),
University of Turku, Turku, Finland, 477-86.
Yarar, E. (2002). The Official Language of Turkish : A Formal and Functional Approach, Unpublished Ph.D. Dissertation,