Models of Linguistic Description

Models of Linguistic
Description





Models and idealization process


The complex nature of language, its bewildering assortment
of cognitive and social elements makes


it difficult to study. The particularities of language, in
fact, distract any linguist because of their


instability. What is needed is an abstraction process that
gets rid of these particularities that are


peripheral and residual. This is not new in the field of
scientific enquiry. Other fields than linguistics


do rely on this process. Chemistry, Physics, etc are
disciplines that relies on abstraction too. This


abstraction process is called idealization.


I shall take idealization to mean the thinking process by
which you abstract particulars from


generalities. In other words, it is as I defined it
elsewhere (Neddar 2004:131) ‘the abstraction of the


formal properties of the language code from the contextual
circumstances of actual instances of use’.


In this process one needs to dissociate himself from the
immediacy of the context and hence be at a


remove from the actual. No matter how details are important,
they are always a hindrance that can


mislead us in our research as they are by their very nature
variable. Stability, it seems to me, is one of


the key elements in any scientific query.


This process involves according to Lyons (1977) three stages:


i. Regulation: Under this head, we can discount all
slips of the tongue, mispronunciations, hesitation


pauses, stammering, stuttering, etc.: in short, everything
that Chomsky attributes to the influence of


such microlinguistically irrelevant factors as memory
limitations, distractions, shifts of attention and


interest, and the malfunctioning of the psychological and
neurological mechanisms involved in


language-behaviour.


ii. Standardization: This is the second kind of
idealization. When we say that two people speak the


same language (e.g. Algerian Arabic), we are, whether we are
aware of it or not, abstracting from all


sorts of systematic differences in the language –systems
which underlie their language-behaviour.


This Algerian Arabic covers many variations and accent. The
Oran variation of AA is quite different


from that of Algiers which in its turn is different from
that of Annaba for instance. Others are


attributable to such factors as sex, age, social status, social
role, professional occupation, many of


which have been described as contextual variables. There is
a sense in which it is true to say that


everyone we normally describe as a native speaker of
Algerian Arabic speaks a different Algerian


Arabic: he has his own language-system, distinct to some
degree in vocabulary, grammar and


phonology. Indeed, every native speaker of Algerian Arabic
speaks many varieties of AA and uses


them in different situations.


It would be absurd to hope to describe, or even to determine,
all these differences within what we


call Algerian Arabic. What the linguist does, in practice, is
to focus on the standard features that are


as the word suggests, shared by all the variations of the
same language, and this is what is meant by


standardization. A good example of that is the language (AA,
of course) used in some Algerian series


such as ‘ El Bidra’, ‘ Ashwak El Madina’ and so on, where
the focus is more on a standard version that


can be understood all over the country and where every
Algerian viewer can recognise himself in it.


iii. Decontextualization: This is the third kind of
idealization that is involved in the process of


abstraction. Spoken utterances of everyday conversation tend
to be heavily context-dependent, as


well as being characterised by errors and other performance-phenomena,
which, we are assuming,


are eliminable by regularization.


To sum up, I would say that the process of
decontextualization is a process that isolates the


sentence (a unit of form, i.e. a competence category) from
its corresponding utterance (a unit of use,


i.e. a performance category), in other terms it is a process
of isolating the contextual elements from


the textual ones.


The notion of idealization leads us to another and no less
important one, that of a model which will


be the concern of my next discussion.


A model is an idealised version of the reality. It is meant
to refer to it without resembling it. Indeed,


it ceases to be so if ever it does. All the unimportant
elements are stripped so that to make apparent


what is essential and which is in normal events not
important (Widdowson 1966).


The purpose of linguistics is to study language by focusing
on the elements that are not apparent


and important. With this in mind, let’s look at the two
major models of linguistic description:


Saussure’s and Chomsky’s.


Saussure’s model


Saussure’s model is based, of course, on idealization. In
search of stability of language Saussure made


a distinction between langue and parole. Langue is an
abstract knowledge which nature is social. It is


likened to a common reference manual book acquired by all
the members of a given speech


communities. Langue is to be thought to represent the
central and important aspect of language.


Parole, on the other hand, is seen to be the executive side
of langue. It is its realisation as an actual


linguistic behaviour. It is individual and thus
heterogeneous in nature which implies that it is elusive


and unstable.


This distinction suggests that langue is stable, however, we
do know that language is a process not a


static product; it varies through time and develops to meet
the need of its users. As a linguist brought


up in the tradition of historical linguistics, Saussure was
well aware of that. This leads him to think


about a second dichotomy: diachronic versus synchronic. So, he
conceives of langue as a crosssection


of language in a particular point of time. He argues that if
language evolves through time


(diachronically); it is fixed at a certain point of time (synchronically).
He calls this cross-section ‘L’ etat


de langue’ (the state of langue). He argues that this can be
likened to the game of chess where the


game stops each time a player makes a move. One, he says, can
study the pattern of the pieces as a


state without taking account of their movements before or
after the move of the player is made.


What we notice here is that there is confusion between
synchrony and stability. To claim that


language is stable whenever we take a synchronic slice, is
to deny the fact that it varies even at that


moment since it is spoken by different generations and
generations generates differences. In fact,


the diachronic change of language is due to its synchronic
variation.


Chomsky’s model


Another comparable distinction to that of Saussure’s is that
of Chomsky who draws a distinction


between competence and performance. Competence, for him, is
that abstract formal knowledge


that native speakers have of their language. Performance, on
the other hand, is its actual realisation


as a linguistic behaviour. Though performance is projected
from competence and refers to it, as


Widdowson (ibid:24) argues, it does not equal or correspond
to it. In fact, we do not always act on


what we know. Our performance does not necessarily reflect
our knowledge. We know more than


we do. This simply because our linguistic behaviour, speech -to
make it simple- is affected by other


factors, some external to language itself. These factors are
considered by Chomsky as being residual,


peripheral and of secondary importance, just as with parole
for Saussure.


Both Sausure’s langue and Chomsky’s performance are viewed
in terms on abstract knowledge,


however the nature of this knowledge differs. The following
comparative table illustrates the


similarities and differences between the two models:





Saussure’s Model


- Parole is peripheral and irrelevant


- Langue: abstract knowledge


- The nature of langue is social, shared by all members of
the community. It is


a common knowledge.


- Langue is determined by membership to of a social
community. You know a particular language because you belong. to a particular
speech community


- What we are interested in is how languages Differ from one
society to another, what is distinctive of languages as a social phenomena





Chomsky’s Model


- Performance is peripheral and irrelevant too


- Competence: abstract knowledge


- The nature of competence is cognitive, psychological;
it is a genetic endowment in each individual.


- Competence is determined by membership to the
human species. You know language because you are human.


- What we are interested in is how individual competences
are similar. What is distinctive about language in general?