LINGUIST List 15.578

Fri Feb 13 2004

Review: General Ling: Halliday (2003)

Editor for this issue: Naomi Ogasawara <>

What follows is a review or discussion note contributed to our Book Discussion Forum. We expect discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in. If you are interested in leading a book discussion, look for books announced on LINGUIST as "available for review." Then contact Sheila Dooley Collberg at


  1. Salvio Mart�n Men�ndez, On Language and Linguistics, Volume 3.

Message 1: On Language and Linguistics, Volume 3.

Date: Thu, 12 Feb 2004 21:34:23 -0500 (EST)
From: Salvio Mart�n Men�ndez <>
Subject: On Language and Linguistics, Volume 3.

AUTHOR: Halliday, M. A. K.
EDITOR: Webster, Jonathan J.
TITLE: On Language and Linguistics, Volume 3
SERIES: The Collected Works of M. A. K. Halliday
PUBLISHER: Continuum International Publishing Group Ltd
YEAR: 2003

Announced at

Salvio Mart�n Men�ndez, Universidad de Buenos Aires,
Universidad Nacional de Mar del Plata.


On Language and Linguistics, the third volume of the Collected Papers
of M. A. K. Halliday, is a selection of eighteen central papers of his
work, including a new paper that serves as introduction called ''On
the 'Architecture' of Human Language''. The eighteen papers are
divided in three parts. The first part is ''Linguistics as
Discipline'' (Chapters 1 to 6); the second one, ''Linguistics and
Language'' (Chapters 7 to 13); the third one, ''Language as Social
Semiotics'' (Chapters 14 to 18)

The introductory paper is not only an excellent updated version of
Halliday's Systemic-Functional Theory, but a point of departure to
read (or read again) the selection that follows as well. This paper is
crucial to understanding Halliday's way of describing, explaining and
interpreting language in use, i.e. language as a social instrumental
tool within a social semiotic frame.

Halliday's work has always been clear. This paper is, maybe, one of
the clearest piece of work he has written. This is, probably, because
he does not want to dazzle. He has just wanted to be clear. And his
clarity is dazzling.

Then he describes a number of central points that will appear
throughout the volume. These points can be summarized as : language as
a system of meaning, i.e. a semiotic system; types of complexity in
language; paradigmatic composition of language; stratification of
language; metafunctions : the grammar at work; syntagmatic
composition; congruent and metaphorical modes of meaning; probability
and instantiation, variation and fuzz.


Chapter One is the oldest paper of the selection. It is called
''Syntax and the consumer'' (1964). Halliday clearly points out that
linguistics and linguists must recognize that language may be
described for a wide range of purposes. Then, he makes a very keen
evaluation of Chomsky's contribution to define the goals of linguistic
theory. He strongly states that one point of view is not ''the'' point
of view. He points out clearly that his orientation is ''primarily
textual and, in the widest sense, sociological'' (1993: 40). It
includes what he calls the ''scale of delicacy'' that involves a
characterization of the special features of the varieties of languages
used for different purposes.

The second paper, ''Grammar, society and the noun'' (1966) lets
Halliday include himself within the sociolinguistic tradition in a
wide sense (Firth, Sapir, Whorf, Hymes). He uses the history of the
uses of noun and the process of nominalization as examples to show how
they have been used to shape reality and how reality, therefore, has
been interpreted out of them.

In the third paper, ''The context of linguistics'', Halliday points
out in an epigrammatic style the main point he wants to make:
''Linguistics became the study of linguistics rather than the study of
language'' (1993: 75). He clearly argues that formal linguistics has
focused on intra-action and that language is social. He needs, then,
to give a broad picture including not only what an ideal speaker
knows, but how a real speaker acts when he/she uses language in
everyday interaction. He affirms: ''society, language and mind are
indissoluble: society creates mind, mind creates society, and language
stands as mediator and metaphor for both these processes'' (1993: 90).

In Chapter Four, ''Ideas about language'' (1977) he begins by saying
''Linguistics may be still a new name; but it is by no means a new
phenomenon''(1993:92). So, he makes an historical account about
language from Panini and Greeks until today. He makes sharp remarks
about Scholastic Tradition, French Rationalist Grammar, Philosophical
and Ethnographic Traditions, Structuralism and Generativism (he
regrets that a dialogue between them should have happened, but it did
not),and Variationism (Labov). Halliday postulates two ways of seeing
language: language as resource and language as rule. He prefers the
first one, because it leads him to semiotics. ''Semiotics is not a
discipline, defined by subject-matter. It is a way of interpreting
things'' (1993: 113).

In Chapter Five, ''Language and the order of nature'', Halliday says:
''Language is as much a product of evolution as we are ourselves: we
did not manufacture it. It is an evolved system, not a designed
system: not something separate from humanity, but an essential part of
the condition of being human'' (1993: 117). He clearly sees two
complementary orders constructed out of it: the social order and the
natural order. Examples of the discourse of physics and oral and
written speech are used to sustain his point.

In Chapter Six, ''New Ways of Meaning: The Challenge to Applied
Linguistics'' (1990), he talks about a special discipline, applied
linguistics, defined according to its content. He clearly establishes
its scope: language teaching, language policy and planning. As
language - he affirms - construes reality, grammar must be based in
language in its every day form; finally, grammar is, in this way, a
theory of experience. The goal of planning a language is a fundamental
task. His metaphorical conclusions cannot be more adequate: ''We
cannot transform language; it is people's acts of meaning that do that
(.) I do not suggest for one moment that we hold the key. But we ought
to be able to write the instructions for its use'' (1993: 171-172).

Chapter Seven, ''A brief sketch of systemic grammar'' (1969), is the
first paper of the second part. In this short article, Halliday points
out that grammar is based on the notion of choice and outlines its six
main features: 1. the system network is the grammar; 2. the
description of a sentence, clause, text may be just the list of
choices that the speaker has made; 3. the options selected by the
speaker are realized as structures; 4. any item may have not just one
structure but many; 5. there are two classes of labelling: according
to structural function (actor, for example) or according to class
(noun phrase, for example) 6. the systemic-grammar has to indicate how
the particular choices made by the speaker are realized in structural

In Chapter Eight, ''Systemic Background'' (1985) Halliday explains the
historical roots of systemic-functional linguistics. He clearly
indicates: ''The theory has evolved in use; it has no existence apart
from the practice of those who use it. Systemic Theory is a way of
doing things'' (1993: 185). He points out the ''broad foundations on
which systemic theory is built'' (1993:186) and he traces his own
intellectual biography in close relation with the theory he has
founded. There are several names; the most important are Firth
(Halliday's teacher), Hjelmslev, Malinowski and Whorf. But he thanks
also Troubetskoy, Benveniste, Whitehead, Wittgenstein, Sapir,
Bloomfield, Fries, Hockett, Harris, Gleason, Pike, Allen, Robins,
Henderson, Whitley and Lamb. He then establishes the main
characteristics of systemic theory (it is an excellent complement of
Chapter Seven). The last characteristic is a good resume: ''Systemic
theory is a way of thinking about language and of working on
language'' (1993: 197).

In Chapter Nine ''Systemic Grammar and the Concept of a Science of
Language'', Halliday clearly establishes the two problems he is going
to face in this paper: the first one is fairly obvious, language; the
second one is not fairly obvious, science. He clearly differentiates
how scientists construct theories from the theories that explain how
science works. He affirms: ''A science of meaning is potentially
rather different from a science of nature, or of society''
(1993:199). Then, he enumerates certain principles and practices that
he thinks are followed by systemic-functional linguistics.

In Chapter Ten, ''Language in a changing world'', Halliday faces
postmodernism and its relation with semiotic systems as constructors
of reality. He clearly points out that modern societies (opposed to
postmodern ones) are still technology driven. But technology has
changed, because it is now the technology of semiosis. He strongly
affirms: ''The linguist who claims to be theory-free is like the
conservative who claims to be non-political: they are both saying , to
be impartial is to leave things as they are - only those who want to
change them are taking sides'' (1993: 223).

Chapter Eleven, ''A recent view of ''Missteps in Linguistic Theory''
(1993), is a review of John E. Ellis's ''Language, Thought and
Logic''. Three are the missteps. They are presented as followed: 1 ''
The assumption that the purpose of language is communication; 2. ''The
assumption that descriptive word like 'square' or 'cat' are simpler
than evaluative words like 'good' and 3. ''the assumption that the
categories of language serve to group like things together'' (1993:
232). Halliday agrees with the picture presented by Ellis, but he
keenly shows that he (and the tradition which he belongs to) has been
doing what Ellis points out that linguistics has to do. Ellis's
problem is to consider that generativism is linguistics. And Mr. Ellis
- Halliday clearly argues - is not the only one to have this problem.

In Chapter Twelve, ''Linguistics as metaphor'', Halliday postulates
that language is a meaning-making system, a property that he calls
semiogenesis. A theory, he argues, is also a semiogenic system and
this applies to all theories (scientific and linguistic theories)
because they create meaning. He describes five features ''from which a
language derives its semogenic power and is enabled to evolve and
function as a self-organizing system'' (1993:250).

In Chapter Thirteen, ''Is the grammar neutral? Is the Grammarian
Neutral?'' (2001), Halliday departs from a provocative phrase said by
Basil Bernstein in the 22nd International Systemic Functional Congress
in Gent in 1994: ''Grammar is never neutral''. Halliday refers to his
biography to establish up to what point he agrees with Bernstein's
phrase. He clearly points out how systemic-functional linguistics
evolved from European linguistics with social and functional
orientation. From this point of view, linguistics ''as (like language
itself) a mode of action, a way of intervening in social and political
processes; and this has remained as a significant motif of work in
systemic functional linguistics'' (1993:273). He says that ''in the
immediate sense the grammar of any language is neutral, in what you
can use it to produce discourse supporting every possible subject
positioning and ideological stance, at another level it is highly
partial: it construes the world from the standpoint of a given moment
in history'' (1993:285). His conclusion is clear: grammars are
neutral, grammarians are not because they cannot.

Chapter Fourteen opens the third and last part of the volume with the
already classical paper ''The functional basis of language'' (1973)
appeared in the Volume 2 of ''Class, Code and Control'' edited by
Basil Bernstein. In this paper Halliday stresses what he understands
as a ''functional approach'' to the study of language. This approach
centers in the social functions of language that clearly determine the
register repertoire of a community. Register or diatypic variation of
language determines the different fields, modes and tenor of discourse
and the system has to be able to accommodate to it. To illustrate his
point, he analyzes a detailed example from child language (Nigel's
language at age nineteen month) where he describes and explains the
different uses of language out of the metafunctions ideative,
interpersonal and textual.

In Chapter Fifteen, ''Towards a Sociological Semantics'' (1972)
Halliday defines what semantics means within the systemic-functional
approach. By stressing the instrumentality of linguistics rather than
its autonomy, he clearly points up: ''Semantics (...) is 'what the
speaker can mean''' (1993:323). And then he specifies: 'can mean' is
one form of 'can do'. An example of the use of language by a mother
for the purpose of controlling the behavior of a child is given.

In Chapter Sixteen, ''The History of a Sentence'', Halliday is
concerned about the history of a semiotic event, that is, an act of
meaning that he defines as follows: '' An act of meaning is a special
kind of semiotic act made of linguistic meaning specifically through
wording''.(1993:355). This is his starting point to speak about the
history of meaning and to show the need that ''any act of meaning must
rest on other such acts that have preceded it and created the
conditions for its occurrence'' (1993:358). To accomplish that four
dimensions of history are needed. He describes them as:
1. Intertextual; 2. Developmental; 3. Systemic, and 4. Intratextual.
He concludes: ''Only if we know what went before an act of meaning can
we reasonably claim to judge its effect on what came after it''

In Chapter Seventeen, ''The act of meaning'' (1999), Halliday says
that the act meaning is a social act, but also a biological, physical
and, finally, semiotic. Then, he faces the problem of linking theory
and instance, not as different things but as two views of the same
phenomenon. His conclusion is clear: ''the full creative power of an
act of meaning arises from the fact that language both construes and
enacts'' (1993:384). Therefore, the system is the text and ''it is in
the act of meaning that the power of language resides; and that is
what makes linguistic systems, in the last resort, subject to the
democratic process'' (1993:389).

The last chapter, the eighteen, is ''On Language in Relation to the
Evolution of Human Consciousness (1995)''. In this paper, Halliday
analyzes the relation between language and the human brain. He
considers that this relationship has to be considered from three
different time scales: evolutionary (how the language-brain evolved in
the species), developmental (how the language-brain develops in each
child) and instantial (how the language-brain is activated in each act
of meaning). He concludes by saying that ''Language is not and outward
and imperfect manifestation of some idealized entity called mind. It
is an evolving eco-semiotic system-&-process, constituting the most
recent phase of evolution of the mammalian brain. Higher-order
consciousness is symbolic consciousness - or better (since
''symbolic'' might still imply the re-presentation of something that
lies beyond) semiotic consciousness; and semiotic consciousness is
another name for meaning'' (1993:429).


All the papers presented in this selection are very important. Each of
them gives an exact picture of the importance of systemic-functional
linguistics as a whole. Halliday always faces the different issues
about language from a broad perspective. He is not fighting to win any
war, but trying to understand how language really works.

Not only are the papers very clear, but they are also well written.
Halliday has style and it is really a pleasure to read him. Each paper
stands as a brilliant lesson of the master.

It is important to noting the excellent work carried by the editor,
Jonathan Webster. One cannot think of a better way of organize the
material. Each part is well arranged and one can clearly how the
different positions have evolved or changed. He has really done a very
good work that deserves to be recognized.

Halliday has always pointed out that we are dealing with everyday
common sense grammar and texts because linguists are trying to explain
what the speakers already know. This collection is the best proof that
he has achieved what systemic-functional linguistics wants to explain.

It is a must for any person interested in language - linguists


Salvio Mart�n Men�ndez is Professor of General Linguistics and
Text Grammar at the Facultad de Filosof�a y Letras of the
Universidad de Buenos Aires and of Linguistics I and II at the
Facultad de Humanidades of the Universidad Nacional de Mar del Plata
(Argentina). He has been working on Pragmatic Discourse Analysis on
different corpora such as Political Discourse, HIV Propaganda
Discourse, and High School Textbooks Discourse.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue