A new approach to register variation: the missing link

06 GimØnez.qxp 20/09/2006 13:49 PÆgina 89
A new approach to register variation: the
missing link
Rosa Giménez Moreno
Universitat de València, Spain
rosa.gimenez@uv.es
Abstract
The purpose of this article is to direct the attention of the specialists towards the notion
of linguistic register in order to promote new systematic methods of analysing its current
span of variation. Despite its still unclear definition and frequent amalgamation with
other varieties, the movement across registers is one of the most effective and frequent
communicative tools that a language has to adapt itself to the diverse private, social and
professional settings. However, this type of variation is one of the most difficult to be
perceived, learnt and correctly used by foreign speakers and, in general, all non-effective
communicators. As this article shows in the first part, the existing theory and models of
analysis need to be reviewed and updated in order to provide clearer distinctions and
more practical parameters of research. The second part of the article proposes a basic
method of approaching current register variation systematically, emphasizing the most
controversial aspects in the analysis of registers and pointing out the need to overcome
these difficulties in order to promote more practical studies, other useful typologies and
new pedagogical materials on this type of variation.
Key words: sociolinguistics, language variation, register, context, communication
Resumen
Nueva perspectiva sobre la variación de los registros lingüísticos: el eslabón
perdido
El objetivo del presente artículo es dirigir la atención de los especialistas hacia la noción
de registro lingüístico con el fin de promover nuevos métodos de analizar los registros
actuales. A pesar de la falta de claridad en su definición y la frecuente confusión con otros
tipos de variación lingüística, el movimiento a través de toda la gama de registros es uno
de los instrumentos comunicativos más efectivos y de uso más frecuente que posee una
lengua para adaptarse a los diversos entornos privados, sociales y profesionales. Sin
embargo, este movimiento es uno de los mecanismos más difíciles de percibir, aprender
IBÉRICA 12 [2006]: 89-109
89

---

06 GimØnez.qxp 20/09/2006 13:49 PÆgina 90
ROSA GIMÉNEZ MORENO
y utilizar correctamente, sobre todo en el caso de hablantes extranjeros y comunicadores
poco efectivos. Como se muestra en la primera parte del artículo, la teoría y los modelos
de análisis existentes necesitan ser revisados para poder generar distinciones internas y
parámetros de investigación más claros. En la segunda parte del artículo se propone un
método básico de análisis que permite la ejemplificación sistemática de los registros
lingüísticos actuales, enfatizando los aspectos más controvertidos que operan en este tipo
de análisis y la necesidad de superarlos para conseguir elaborar estudios más prácticos,
tipologías más útiles y nuevos materiales pedagógicos.
Palabras clave: sociolingüística, variación lingüística, registro, contexto, comunicación
Introduction
Since variability is inherent in human language and most of this variation is
systematic, from the 70s linguistics has focused on language use and the systematic
ways in which users exploit linguistic resources. From this perspective, many studies
have approached linguistic variation analysing linguistic features such as hedging or
modality, language dimensions or functions which make a discourse more
informative or narrative, and genres such as every day conversation or newspaper
articles, just to mention a few examples; most of them compiled in famous corpora
such as the Michigan Corpus of Academic Spoken English (MICASE), the British National
Corpus or the International Corpus of English (Biber, 1995, 1998; Reppen et al., 2002).
Nevertheless, as most specialists admit (Reppen et al., 2002: VII), “analyzing the
influence factors on linguistic variation present difficult methodological challenges.”
Among these difficulties, we may include carrying empirical analyses of natural texts,
collecting texts from a wide range of speakers and contextual factors, and
establishing clear differences among all the varieties interacting in a speech event.
These three obstacles, and in particular the third one, have become the key stone of
the avoidance by many authors of questionable concepts such as the concept of
“communicative register.” There have been important well-known attempts to define
the term and thanks to these proposals we can have a better idea of the uniqueness
and specificity of the concept. However, as it can be observed in real practice and
throughout most published research on language variation, many specialists still use
the word “register” in an unspecific and unclear way, in many cases exchanging the
word randomly with the other types of variation, such as genre, style, etc. This switch
is probably done in the belief that it might be better understood in that way, but
sometimes it is because what they are actually investigating or making reference to is
not register itself but another of these other language varieties. In practice, it seems
apparent that a unified and widely accepted clear use of register does not exist as a
IBÉRICA 12 [2006]: 89-109
90

---

06 GimØnez.qxp 20/09/2006 13:49 PÆgina 91
A NEW APPROACH TO REGISTER VARIATION
type of language variation different from the other varieties –as it happens, for
example, with the concept of genre and idiolect. If language variation as a whole is
seen as a chain of specific choices made by users according to diverse communicative
aspects brought into play, register variation seems to be a “missing link” inside this
chain, without a clearly defined position, nature and usage.
Taking these factors into account the purpose of this paper is, firstly, to direct the
attention of specialists towards the notion of register, its still unclear definition,
current random use and frequent amalgamation with other varieties; and secondly, to
emphasize the need of facing the main controversial areas and approaching the
subject from new perspectives which help to understand its main defining
components, its specific behaviour in a particular language or across languages, its
specific influence in daily communication and its complete scale of internal variation.
Register analysis: a critical overview
Linguistic theory only counts on a few theoretical approaches to carry out systematic
research into current register variation and, as this paper will point out, even these
well-known approaches do not seem to cast enough light on the subject to promote
practical studies, useful typologies and effective pedagogical materials on this type of
variation. For example, Coultas’ treatise, Language and Social Contexts (2003), clearly
gives information on variation regarding power, gender and regional talk but deals
with register variation only by mentioning some “formality” parameters (Coultas,
2003: 9). It seems that the scale of formality is still the most useful way of
approaching, using and assessing this type of variation. Obviously, although the
concept of register can be included inside the general field of language variation, as
underlined by Biber and Finegan (1994), its true nature places the analyst in a highly
delicate research area located in the junction of important fields such as
sociolinguistics, language in use, pragmatics, applied linguistics, discourse and
conversation analysis, speech act theory, code switching, institutional linguistics and
corpus linguistics, among others. It is not easy to study a concept with such a huge
range of research levels and with such a fixed dual conceptualization –which have, in
fact, become almost fossilized in daily practice– therefore restricting its variation to
the broad swinging usage between the concepts of formality and informality
(Giménez Moreno, 1994).
A “register,” from a general social perspective, can be seen as the adequate manner
of expressing your message depending on the social situation where that message is
communicated; a communicative frequency or wavelength chosen by the individual
IBÉRICA 12 [2006]: 89-109
91

---

06 GimØnez.qxp 20/09/2006 13:49 PÆgina 92
ROSA GIMÉNEZ MORENO
from the whole scale of possibilities given by a language to express a certain idea
properly within its context. Contextual variation started to be an interesting field of
research and study at the beginning of the 20th century together with all the emergent
anthropological and ethnological descriptive analyses of the time. Following the work
of pioneers such as Malinowski (1923) and Firth (1935) on the importance of the
context in language variation, the term “register” was suggested by Reid in 1956 with
reference to the capacity of human language to adapt itself to different social
situations. With this proposed terminology, Reid was pointing out that this adaptation
operates within a vertical or transversal mode/axis with regard to the other types of
language variation that operate in a horizontal mode/axis. A few years later, this split
or bifurcation was specified by Halliday et al. (1964) establishing a distinction
between dialectal variation dependant on the user’s characteristics and register
variation dependant on the circumstances of the usage. At that time a few interesting
typologies appeared, being still useful to establish primary differentiations among
registers.
An approach of special interest is the pioneer classification proposed by Joos (1961).
Although this author still mixed the terms “style” and “register” and did not provide
a detailed linguistic account of the main differences among the proposed styles, he
transcended the traditional dichotomy between formal and informal language and
established correlations between five types of “style” (frozen, formal, consultative,
casual, and intimate) and the age (senile, mature, teenage, child and baby), the breadth
(genteel, puristic, standard, provincial, and popular) and the social responsibility of
the speaker (best, better, good, fair, and bad). Through this perspective, Joos pointed
out that the choices dependent on the user and on the use are closely related and can
interfere with each other; also that the relationship between the interlocutors and the
purpose of the message are key distinctive parameters when distinguishing among
registers. According to this approach, the “consultative style” is the most neutral and
from which the other four registers depart. This style is used with strangers and
unfamiliar interlocutors; its main function is interaction and the speakers have to
provide as much information as possible, not assuming any common ground and
ensuring that the basic content and norms are shared. On the informal side, the
author places the “casual style” on a first level, being the style used with known or
familiar interlocutors who share the same communicative norms and mechanisms.
He places the “intimate style” on a second level of informality being used inside
exclusive groups of speakers who have private or personal ties and do not need to
refer to any general public information. On the other side of the consultative style,
there is also a first level, where Joos (1961) places the “formal style,” which is not
participative or interactive but mainly informative expressed in a linear, explicit and
IBÉRICA 12 [2006]: 89-109
92

---

06 GimØnez.qxp 20/09/2006 13:49 PÆgina 93
A NEW APPROACH TO REGISTER VARIATION
elaborate manner. Finally the “frozen style” is placed on the highest level of
formality, becoming the style of creativity, declamation and printed expression. These
labels and general parameters on register variation proposed by Joos in the 60s are
still useful in our current practice but they have not been revised, further developed
and made applicable to specific contexts inside the whole spectrum of daily
communication.
Despite the efforts of these authors and other specialists (Barber, 1962; Gregory,
1967) in favour of register analysis, at that time, the relevance of the research on
social variation was shared –and often shaded– by that dedicated to other many social
and dialectal varieties which were targeted by the most important linguists of those
decades. In the mid-60s social situational variation was examined by Ferguson (1959,
1964), on diglossia and baby varieties respectively; Brown and Gilman (1960), on
language and power; Leech (1966), on press advertising and poetry; Crystal and Davy
(1969), on conversation, newspapers, legal and religious language; Gumperz and
Hymes (1972), on argots in Asian villages; Labov (1972a), on urban and social
dialects; and Trudgill (1974), on social differentiation of English; among many
others. Most of these studies identified registers with styles (Labov, 1972a,b; Trudgill,
1974); with some rare exceptions such as Fisher (1958) who already distinguished
three levels of formality in his interviews. According to Reppen et al. (2002), the
factors attributed to this neglect of register research were, on the one hand, the fast
development of descriptive functional linguistics and the prevailing status of
spoken/conversational language (viewing the analyses of written varieties as targets
of literary and rhetorical concern); and, on the other hand, “the discounting of the
importance of register differences in a speech community, and the extent and social
importance of the linguistic differences associated with those register ranges”; adding
to these factors “the lack of a paradigmatic, or even dominant, methodological and
theoretical framework for the analysis of registers” (Reppen et al., 2002: 6).
Nevertheless, register studies were not totally ignored and the term was often used in
the studies carried out by some of the most important specialists of this period
(Ferguson, 1964; Gumperz & Hymes, 1972; Labov, 1972a,b). However, from a
practical point of view, the problem was that –as Benesh (2001: 5-6) mentions in her
chronological review of EAP– most “register analysis” carried out in the 60s were
lexical analyses on a specific topic or discipline such as “scientific English” or
“Business English.” She gives as an example, Ewer and Latorre’s Course in Basic
Scientific English (1969). These authors analysed three million words within the scope
of modern scientific English ranging from popular writing to learned articles and
graded the words according to frequency and complexity. From that study,
pedagogical material appeared but, as it has been pointed out, all these materials were
IBÉRICA 12 [2006]: 89-109

---

06 GimØnez.qxp 20/09/2006 13:49 PÆgina 94
ROSA GIMÉNEZ MORENO
about a “discipline” or a “topic” expressed in what might be called a “neutral or
standard professional register.” This register, however, became very helpful for the
learning of specific vocabulary and grammatical structures related to a specific
professional field but not for the distinguishing and learning of the register variation
that operates in the communication related to that topic or field. For example, the
process of water condensation and raining –“scientific English”– can be described to
a group of university students sitting in a lecture theatre (institutional/academic
register) but also to our own child lying in the garden (intimate register). The topic
will be the same but the register in both cases might be quite different: in the first
case, the speaker –the lecturer– will tend to use impersonal expressions, elaborate
connectors, specific technical vocabulary, Latin terms, nominalizations and restricted
body language; whereas in the second case, the speaker –the mother/father– might
use attitudinal expressions, personalization, contractions or “fast” language, idiomatic
expressions, figurative and metaphoric language, basic connectors and free body
language; among many other choices.
In the 70s and 80s, specialists (Hymes, 1972; Gregory & Carroll, 1978; Halliday, 1978)
concentrated on the limits of the concept of speech situation regarding those fields
of linguistics highly dependant on its influence: speech acts, inference and
presupposition, topic and comment, rhetorical devices such as metaphor, irony and
allusion, etc. Together with all these new insights, more register studies emerged
analysing the “register” of other specific disciplines, mainly professional, but most of
them were still topic-centred lexical contributions. As Fowler (1996: 190) pointed out,
the limitations among these “registers” were so confusing that it was very difficult to
draw any classification. Despite this general tendency, three new important aspects
were underscored with regard to register variation:
(a) the need to analyse register from a wider perspective which allowed
embracing other language varieties and including other communicative
parameters (Hymes, 1984; Martin, 1985);
(b) the importance of establishing some clearer set of differences between
register analysis and genre analysis (Ventola 1984; Couture, 1986; Swales,
1990); and
(c) the key connection between register variation and the “verbal repertoire”
of each language community: “no human being talks the same way all the
time […]. At the very least, a variety of registers and styles is used and
encountered; […] each language community has its own system of
registers” (Ure, 1982: 5).
IBÉRICA 12 [2006]: 89-109
94

---

06 GimØnez.qxp 20/09/2006 13:49 PÆgina 95
A NEW APPROACH TO REGISTER VARIATION
This theory clearly pointed out that registers should be considered verbal repertoires
which depend on the specific language and community conventions (contextual,
social and cultural conventions), not so much on individual conventions, clearly
differentiating registers also from idiolects and communicative styles. Another
essential contribution of these years in the distinction and classification of registers
was the need to concentrate on the marked forms of each repertoire. All speakers
have their own repertoires and know different ways of expressing the same
information but, in a certain context, there is one which predominates over the
others as the “marked form”: "These marked forms of language constitute the
registers of a linguistic repertoire" (Finegan & Besnier, 1989: 429). However, as
McIntosh (1963) had warned 20 years earlier, these repertoires and marked forms
depend on the social conventions and the perception of communicative suitability
and appropriateness developed by the speech community.
Despite the relevance of these theoretical insights and their potential to develop
diverse methodological approaches to register variation, the general practice still
tended to identify register with topic/discipline, genre or style, always relating the
notion of register to the scale of formality, and most of the specific register theory
along this period evolved under the functional systemic approach to register variation
led by Halliday (Halliday et al., 1964; Gregory & Carroll, 1978; Halliday, 1978, 1980)
based on the three well-known parameters: “field” (topic or activities involved),
“tenor” (participants’ role relations) and “mode” (communication channels). They
also established the correlation between each of these three components and the
three main meta-functions of language in a social context: conceptual (field),
interpersonal (tenor) and textual (mode) (Halliday, 1978). This perspective –highly
theoretical– integrated and justified as register-related analyses the many previous
studies on diverse disciplines (topic or field-centred analyses), and included within it
the notion of register variation, not only the increasing studies on written and spoken
communication, (mode) but also all those studies on variation related to role
relationships (tenor), such as doctor-patient, teacher-pupil, vendor-customer, and so
on (Sinclair & Coulthard, 1975; Burton, 1980; Ventola, 1987; among many others).
This wide ranging and all-embracing perspective of register led to the current
situation in which either almost all can be included within register variation or hardly
anything can really be attributed to register variation.
Obviously most of these studies on specific syntactic or grammatical features (e.g.
passive voice), on functions (e.g. informing), on lexical characteristics (e.g.
nominalizations) or on particular genres (e.g. letters and brochures) also offered
subsequent insights into the use of register variation and contributed to obtaining a
grasp of the magnitude of its coverage. However, this partial perspective does not
IBÉRICA 12 [2006]: 89-109
95

---

06 GimØnez.qxp 20/09/2006 13:49 PÆgina 96
ROSA GIMÉNEZ MORENO
allow us to see how all the communicative aspects –syntax, functions, lexis, genres,
etc.– vary together to move from one register to another, which constitutes the
“keystone” to classifying registers and identifying repertoires. On the other hand, the
studies on register variation from this systemic perspective offered some theoretical
frameworks exemplified with linguistic and communicative features associated to a
few differentiated registers –mainly professional registers (see Ghadessy, 1988)– but
did not help to establish a practical clear division between registers and other close
types of variation, these frameworks being very difficult to apply to all registers.
Probably the search for an alternative, a distinctive theory to analyze registers, was the
initial purpose of Biber’s work in the 80s. Biber (1986, 1988) and his colleagues started to
develop a new “multidimensional” (MD) method of register analysis based on the four
most important “dimensions” which differentiate the most frequently used registers:
(a) involved versus informational production;
(b) narrative versus non-narrative concern;
(c) elaborated reference versus situation-dependent reference; and
(d) impersonal versus non-impersonal style.
Coinciding with Crystal (1991), Biber and Finegan (1994) defined a register as a
“linguistic difference that correlates with different occasions of use”; however, they
added the following warning: “in addition to the term register, the terms genre, text
type and style have been used to refer to language varieties associated with situational
uses” (Biber & Finegan, 1994: 4). They distinguished these terms from the notion of
dialect associated with groups of users, but not clearly amongst the four quoted
terms themselves. This is why their work –despite the clarity of the title of the book
(Sociolinguistic Perspectives on Register) and the high quality of the research included–
contains several articles where there is no clear distinction between registers, genres
and styles; finding expressions such as “the register of coaching,” “Personal ads
register” or “Diglossia as register variation,” together with other articles mainly about
styles. This mismatch in how a register may be considered with regard to other types
of language usage keeps the sensation of certain confusion among the specialists
themselves; a sensation which tends to increase dependant upon the specific set of
heterogeneous dimensions proposed by the authors to study this variation: a mixture
of functions (e.g. informational and narrative function) and linguistic features usually
related to the user’s style (e.g. impersonal style).
This unclear notion of register can be observed in many other works; for example,
Schiffrin et al. (2001), Handbook of Discourse Analysis, which included very interesting
IBÉRICA 12 [2006]: 89-109
96

---

06 GimØnez.qxp 20/09/2006 13:49 PÆgina 97
A NEW APPROACH TO REGISTER VARIATION
information about “register variation” in several chapters –mainly following Biber
and Conrad’s (2001) MD perspective– where the authors considered genres such as
telephone conversations, face to face conversations, spontaneous speeches, public
speeches, general fiction, professional letters, biographies, press reviews and official
documents as registers. Something similar happened in Geisler (2002) regarding
debates and trials.
Although the MD method is basically functional and based on the analysis of genres
rather than registers, it is a highly valuable contribution to the study of registers, not
only due to the amount of theory, suggestions, data and examples provided on
language variation, but also to some insights into the distinctive nature of this specific
type of variation. For example, Schiffrin et al. (2001: 191) specified that the main
parameters that had an influence on the choice of a certain register were, on the one
hand, the communicative purpose and the physical relationship between addressor
and addressee, and on the other, the production circumstances and setting. In
contrast with Halliday’s (1978, 1980) theory, this notion does not include the mode
and the field of communication as distinctive parameters in register variation, which
brings the discussion back into the pioneer area of defining the basic parameters
which intervene in the distinction of registers from other types of variation.
Differentiating register from other closely related linguistic
concepts
This need to update and re-elaborate the foundations on register variation can be
observed in contemporary authors such as Davies (2005). In his work on Varieties of
Modern English, Davies analysed what he called “contextual variation” and, although
he followed Halliday’s theory about mode, tenor and field and exemplified this theory
with some genres (cooking recipes and literary novels), his definition of register
specified clearly the two main components of a register: the contextual situation and
the roles of the participants. He started his approach by including as relevant aspects
all the components of a speech situation listed by Hymes in 1972: setting,
participants, ends, act, key, instrumentalities, norms and genre; but eventually he
pointed out as distinctive two basic components: “register is a variety of language
most likely to be used in a specific situation and with particular roles and statuses
involved” (Hymes, 1972: 112).
Therefore, although many different components interact in all types of social
variation at the same time, there is a distinctive correlation between them and the
three main types of contextual variation (register, genre and style):
IBÉRICA 12 [2006]: 89-109
97

---

06 GimØnez.qxp 20/09/2006 13:49 PÆgina 98
ROSA GIMÉNEZ MORENO
(a) the established norms or conventions of the setting and the roles of the
participants decide the type of register to be used (for example,
ceremonial or ritual register);
(b) inside that specific register, the participant’s ends (purpose and function),
the chosen instrumentalities (channels) and the act (sequence and form of
expression) of the communicative event decide the type of genre (for
example, a sermon); and, finally,
(c) the key (mood and manner) –among others– decide the type of style (for
example, ironical).
Considering this telescopic dimension of the language variation Martin (1985) already
observed that the language is expressed through registers and registers through genres;
and following Martin, Carrillo Guerrero (2005: 25) also considered registers on a
rhetorical level above all typologies of genres, texts or discourses, communicative
codes and styles. He distinguished clearly registers from genres, as he exemplified
through the text of the following postcard: “Here for a week with my sister. Been
trying out my German. Lesley,” (Carrillo Guerrero, 2005: 3). This author pointed out
that this text belonged to the genre of post cards, but its linguistic peculiarities were
determined by its register. The ellipsis of the personal pronoun “I,” of the verbs “have
been/have,” of the heading “Dear, etc.” and of the ending “Love, etc.” correlated with
an informal communicative situation of shared trust and knowledge between the
interlocutors which differences it from other post cards of its same genre.
If we take as an example the macro-genre of spontaneous daily conversation and
look at two different conversations: a conversation with our boss in his office (small
meeting) and a conversation with our partner in bed (pillow talk). Regarding register,
these two conversations will be very different and, as we see, the main variable will
be the relationship between the interlocutors (their intentional and factual roles) and,
as a secondary very important variable, the communicative setting (the norms,
restrictions and appropriateness of where and when). Therefore, in practical terms
these are the two basic variables which define the concept of register and on which
extensive register analysis depends. Obviously, all the other variables interact at the
same time and have an influence on the type of language produced. This is why, apart
from this first basic differentiation of registers from genres (e.g. conversation, letter,
report, novel, lecture. etc.) and styles (e.g. ironical, distant, elaborate, plain, etc.), on
defining register variation other specifications are also crucial.
In the last ten years, several specialists approached the boundaries of the concepts of
register, genre and style, and most of them seemed to agree on pointing out the
IBÉRICA 12 [2006]: 89-109
98

---