Introduction
A rough guide to Cognitive Linguistics
Dirk Geeraerts
So this is the first time you visit the field of Cognitive Linguistics, no? You may
need a guide then. Sure, when you move through the following chapters of this
volume, you get to see a top twelve of sights that you should not miss: a delightful
dozen of articles written by authorities in the field that each introduce one of the
conceptual cornerstones of the theoretical framework of Cognitive Linguistics.
Still, to give you a firm reference point for your tour, you may need some initiation
to what Cognitive Linguistics is about. That’s what the present chapter is for:
it provides you with a roadmap and a travel book to Cognitive Linguistics. It’s
only a rough guide, to be sure: it gives you the minimal amount of background
that you need to figure out the steps to be taken and to make sure that you are
not recognized as a total foreigner or a naïve apprentice, but it does not pretend
to supply more than that.
To understand what you may expect to find in this brief travel guide, we need
to introduce one of the characteristic ideas of Cognitive Linguistics first – the
idea, that is, that we should not just describe concepts and categories by means
of an abstract definition, but that we should also take into account the things that
the definition is about, if we are to achieve an adequate level of knowledge. Take
birds: you can define birds as a certain type of animal with certain characteristics
(like having wings, being able to fly, and being born from eggs), but if you want
to get a good cognitive grip on what birds are, you will want to have a look at
some typical birds like robins and sparrows and doves, and then maybe also at
some less typical ones, like chickens and ostriches.
It’s no different when you are dealing with linguistic theories. You have to
know about the scientific content of the theory, that is to say, the abstract definition
of the approach: the topics it deals with, the specific perspective it takes,
and the observations it makes. But you also have to know about the sociology
of the theory: the people it involves, the conferences where they meet, the channels
in which they publish. Introductions to linguistics tend to focus on the first
perspective only, but the present guide will take the second into account just as
much as the first.
2 Dirk Geeraerts
1. What is so special about this place?
Theories in linguistics tend to be fairly insular affairs: each theoretical framework
tends to constitute a conceptual and sociological entity in its own right, with
only a limited number of bridges, market places or even battlegrounds shared
with other approaches. Cognitive Linguistics, when considered in the light of
this metaphor, takes the form of an archipelago rather than an island. It is not
one clearly delimited large territory, but rather a conglomerate of more or less
extensive, more or less active centers of linguistic research that are closely knit
together by a shared perspective, but that are not (yet) brought together under
the common rule of a well-defined theory. The present volume contains an introduction
to twelve fundamental parts of that theoretical conglomerate – a tour of
twelve central islands, if you wish: Cognitive Grammar, grammatical construal,
radial network, prototype theory, schematic network, conceptual metaphor, image
schema, metonymy, mental spaces, frame semantics, construction grammar, and
usage-based linguistics.
We will define in a moment what links hold these concepts together and why
each of them separately is important, but at this point, the chief thing is to realize
that there is no single, uniform theoretical doctrine according to which these
research topics belong together: Cognitive Linguistics is a flexible framework rather
than a single theory of language. From the point of view of category structure (one
of the standard topics for analysis in Cognitive Linguistics), this recognition is
again one way in which Cognitive Linguistics illustrates its own concepts. As we
mentioned a moment ago, Cognitive Linguistics emphasizes the fact that defining
a category may involve describing some of its principal members rather than
just giving an abstract definition. But it also stresses that the abstract definition
need not consist of a single set of defining characteristics that belong uniquely
and distinctively to that category. Think of birds again: when we describe the
features of birds, we soon notice that the features we would like to think of as
definitional for birds are not shared by all members of the species: we may even
find birds like the penguin or the kiwi, that have no wings to speak of, cannot
fly, and don’t have feathers but that are rather covered with some kind of fluff. In
such cases, we say that a category has a family resemblance structure: different
types of birds resemble each other like the members of a family would, but there
is no single set of attributes that necessarily shows up in all the members of the
family. Again, it is no different with a linguistic framework like Cognitive Linguistics:
it constitutes a cluster of many partially overlapping approaches rather
than a single well-defined theory that identifies in an all-or-none fashion whether
something belongs to Cognitive Linguistics or not.
Then again, the recognition that Cognitive Linguistics has not yet stabilized
into a single uniform theory should not prevent us from looking for fundamental
Introduction: A rough guide to Cognitive Linguistics 3
common features and shared perspectives among the many forms of research that
come together under the label of Cognitive Linguistics. An obvious question to
start from relates to the ‘cognitive’ aspect of Cognitive Linguistics: in what sense
exactly is Cognitive Linguistics a cognitive approach to the study of language?
Terminologically speaking, we now need to make a distinction between Cognitive
Linguistics (the approach represented in this reader), and uncapitalized
cognitive linguistics – referring to all approaches in which natural language is
studied as a mental phenomenon. Cognitive Linguistics is but one form of cognitive
linguistics, to be distinguished from, for instance, generative grammar and
many other forms of linguistic research within the field of cognitive science.
What, then, determines the specificity of Cognitive Linguistics within cognitive
linguistics?
There are a number of characteristics that need to be mentioned: one basic
principle that is really, really foundational, and four tenets that spell out this
fundamental notion. The foundational point is simply that language is all about
meaning. As it says in the Editorial Statement of the very first issue of the journal
Cognitive Linguistics, published in 1990, this approach sees language ‘as an
instrument for organizing, processing, and conveying information’ – as something
primarily semantic, in other words. Now, it may seem self-evident to you that
a ‘cognitive’ approach to language focuses on meaning, but if you are familiar
with generative grammar (i.e. Chomskyan linguistics), you will know that this
is a theory that thinks of language primarily in formal terms: as a collection of
formal, syntactic structures and rules (or constraints on such structures and rules).
And generative grammar is definitely also a ‘cognitive’ conception of language,
one that attributes a mental status to the language. So we have to be careful with
the term cognitive in Cognitive Linguistics. It does not only signal that language
is a psychologically real phenomenon (and that linguistics is part of the cognitive
sciences), but also that the processing and storage of information is a crucial
design feature of language. Linguistics is not just about knowledge of the language
(that’s the focus of generative grammar), but language itself is a form of knowledge
– and has to be analyzed accordingly, with a focus on meaning.
Conversely, Cognitive Linguistics is not the only linguistic approach focusing
on meaning: there are diverse forms of functional approaches to language that
go in the same direction. And further, formal semantics is clearly a semantically
oriented approach as well. It lies beyond the scope of this introduction to provide a
systematic comparison with these other semantic approaches, but you will certainly
be interested in what is particular about the way in which Cognitive Linguistics
deals with meaning. So that brings us to the four specific characteristics that we
announced earlier: each of them says something specific about the way Cognitive
Linguistics thinks about meaning. (By the way, the captions we use to introduce
the features may sound formidable, but don’t worry: an explanation follows.)
4 Dirk Geeraerts
Meaning is not just an objective reflection of the outside world, it is a way of shaping
that world. You might say that it construes the world in a particular way, that it
embodies a perspective onto the world. The easiest way to understand the point is
to think of spatial perspectives showing up in linguistic expressions, and the way
in which the same objective situation can be construed linguistically in different
ways. Think of a situation in which you are standing in your back garden and you
want to express where you left your bicycle. You could then both say It’s behind
the house and It’s in front of the house. These would seem to be contradictory
statements, except that they embody different perspectives.
In the first expression, the perspective is determined by the way you look: the
object that is situated in the direction of your gaze is in front of you, but if there
is an obstacle along that direction, the thing is behind that obstacle. In this case,
you’re looking in the direction of your bicycle from the back garden, but the house
blocks the view, and so the bike is behind the house.
In the second expression, however, the point of view is that of the house: a
house has a canonical direction, with a front that is similar to the face of a person.
The way a house is facing, then, is determined by its front, and the second
expression takes the point of view of the house rather than the speaker, as if the
house were a person looking in a certain direction. Such multiple perspectivizations
(and not just spatial ones!) are everywhere in the language, and Cognitive
Linguistics attempts to analyze them.
Meanings change, and there is a good reason for that: meaning has to do with
shaping our world, but we have to deal with a changing world. New experiences
and changes in our environment require that we adapt our semantic categories
to transformations of the circumstances, and that we leave room for nuances and
slightly deviant cases. For a theory of language, this means that we cannot just
think of language as a more or less rigid and stable structure – a tendency that is
quite outspoken in twentieth century linguistics. If meaning is the hallmark of
linguistic structure, then we should think of those structures as flexible. Again,
we don’t have to look far for an example. Think back to what we said about birds:
there is no single, rigid set of defining features that applies to all and only birds,
but we have a flexible family resemblance structure that is able to deal with marginal
cases.
If meaning has to do with the way in which we interact with the world, it is natural
to assume that our whole person is involved. The meaning we construct in
Introduction: A rough guide to Cognitive Linguistics 5
and through the language is not a separate and independent module of the mind,
but it reflects our overall experience as human beings. Linguistic meaning is not
separate from other forms of knowledge of the world that we have, and in that
sense it is encyclopedic and non-autonomous: it involves knowledge of the world
that is integrated with our other cognitive capacities. There are at least two main
aspects to this broader experiential grounding of linguistic meaning.
First, we are embodied beings, not pure minds. Our organic nature influences
our experience of the world, and this experience is reflected in the language we
use. The behind/in front of example again provides a clear and simple illustration:
the perspectives we use to conceptualize the scene derive from the fact that our
bodies and our gaze have a natural orientation, an orientation that defines what
is in front of us and that we can project onto other entities, like houses.
Second, however, we are not just biological entities: we also have a cultural
and social identity, and our language may reveal that identity, i.e. languages may
embody the historical and cultural experience of groups of speakers (and individuals).
Again, think of birds. The encyclopedic nature of language implies that
we have to take into account the actual familiarity that people have with birds: it
is not just the general definition of bird that counts, but also what we know about
sparrows and penguins and ostriches etc. But these experiences will differ from
culture to culture: the typical, most familiar birds in one culture will be different
from those in another, and that will affect the knowledge people associate with
a category like ‘bird’.
The idea that linguistic meaning is non-autonomously integrated with the rest
of experience is sometimes formulated by saying that meaning is experientially
grounded – rooted in experience. The experiential nature of linguistic knowledge
can be specified in yet another way, by pointing to the importance of language
use for our knowledge of a language.
Note that there is a lot of abstract structure in a language: think for instance
of the pattern Subject – Verb – Direct Object – Indirect Object that you find in a
sentence like Mary sent Peter a message. In many languages, such structures are
not directly observable: what we do observe, i.e. what constitutes the experiential
basis for our knowledge of the language, is merely a succession of words (and even
that is not entirely without problems, but let’s pass over those). So the question
arises: how does this more concrete level of words relate to the abstract level where
you find functional categories like Subject and Direct Object? In more traditional
terms, the question reads: how does the lexicon relate to the syntax?
But if we think of grammatical patterns as having an experiential basis in
concrete, observable strings of words, there is yet another step we have to take:
the ‘observable strings of words’ do not exist in the abstract; they are always part
6 Dirk Geeraerts
of actual utterances and actual conversations. The experience of language is an
experience of actual language use, not of words like you would find them in a
dictionary or sentence patterns like you would find them in a grammar. That is
why we say that Cognitive Linguistics is a usage-based model of grammar: if
we take the experiential nature of grammar seriously, we will have to take the
actual experience of language seriously, and that is experience of actual language
use. Again, from the point of view of mainstream twentieth century linguistics,
that is a fairly revolutionary approach. An existing tradition tended to impose
a distinction between the level of language structure and the level of language
use – in the terms of Ferdinand de Saussure (generally known as the founder of
modern linguistics), between langue and parole. Generally (and specifically in
the tradition of generative grammar), parole would be relatively unimportant: the
structural level would be essential, the usage level epiphenomenal. In a usagebased
model that considers the knowledge of language to be experientially based
in actual speech, that hierarchy of values is obviously rejected.
2. What does the tour include?
You are right, of course: the first exploration of Cognitive Linguistics in the previous
section remains somewhat superficial and abstract. You now have a general
idea of what type of scenery to expect in the Cognitive Linguistics archipelago,
but you would like to get acquainted with the specific islands, i.e. you now know
what the overall perspective of Cognitive Linguistics entails, but you hardly
know how it is put into practice. In this section, we will have a look at the twelve
basic concepts that are introduced by the dozen articles in this collection, and
we will show how these concepts relate to the overall picture that was drawn in
the previous pages.
As a preliminary step, let us observe that each of the characteristics that we
discussed earlier defines a number of specific questions for Cognitive Linguistics.
The perspectival nature of meaning raises questions about the specific mechanisms
of construal present in a language: what kinds of semantic construal, imagery,
conceptual perspectivization do languages implement? The dynamic nature of
meaning raises questions about the process of meaning extension: what are the
mechanisms of semantic flexibility, and how do the various readings of a linguistic
expression relate to each other? The encyclopedic nature of meaning raises questions
about the interdisciplinary links of language to the other cognitive capacities:
to what extent are the cognitive mechanisms at work in natural language shared
by other cognitive systems? And the usage-based nature of meaning raises questions
about the relationship between syntax and lexicon, and the acquisition of
language: what kind of experience do children need to learn a language?
Introduction: A rough guide to Cognitive Linguistics 7
These questions are illustrated in various ways by the articles in the collection,
but before we can make that explicit, we need to introduce the articles separately,
and say something about the way in which they are grouped together. Roughly,
there are four groups of concepts and articles, corresponding to the four features
that we identified before. The following pages pay specific attention to the logic
behind the basic concepts that we introduce: why is it that these concepts are so
important to Cognitive Linguistics? What you should see, in particular, is how
they turn the fundamental features that we discussed in the previous section, into
a concrete practice of linguistic description. Reading through the following pages
will give you an initial idea of what you can expect in the volume, but of course,
until you get there yourself, you will never really know what it is about.
2.1. The perspectival nature of grammar
The first two concepts, and ,
illustrate the overall organization of a grammar that focuses on meaning. If conceptual
perspectivization is the central function of a grammar, the typical formal
categories of grammatical description (like word classes or inflection) will have to
be reinterpreted from a semantic point of view. In the context of Cognitive Linguistics,
two authorities in particular are systematically exploring these phenomena:
Ronald W. Langacker, and Len Talmy. The two initial papers in the volume will
introduce you to the thought of these two major thinkers – towering figures in
the context of Cognitive Linguistics, who have both provided the approach with
some of its basic vocabulary.
Cognitive Grammar is the specific name that Langacker uses for his theory of
language. The paper included in the present volume originates from 1986, but is
here reprinted in the form in which it was published in 1990. It specifies a number
of basic features of Cognitive Grammar that are still valid, and that form an
interesting backdrop for the rest of the articles in the present collection. Langacker
starts off with the very idea of a perspectival grammar – although he uses a slightly
different terminology: he talks about grammar as conceptualization and imagery.
He introduces a number of high level general features of grammatical ‘imagery’
(profiling, specificity, scope, salience) and then tackles the key question how to
build a descriptive framework for a grammar that starts from the assumption,
simplistically, that language is meaning and that meaning is conceptualization.
Central to his answer is the idea that a grammar is not built up out of grammatical
rules on the one hand and a lexicon on the other (the idea that you traditionally
find in generative grammar). Rather, a grammar consists of ‘symbolic
8 Dirk Geeraerts
units’, where a symbolical unit is a conventional pairing of a form and a meaning.
You can obviously think of lexical items here, but symbolic units can be more
abstract than that, like when nouns are claimed to instantiate the abstract notion
‘thing’, and verbs the abstract notion ‘process’. Given that you take the notion of
symbolic unit as the basis for a grammar, there are two questions that immediately
crop up, and Langacker does not fail to address them.
First, what could be the notional, conceptual characterization of abstract entities
like word classes? What do we mean when we say that the meaning of nouns
is ‘thing’ and that of verb is ‘process’? On conceptual grounds, Langacker distinguishes
between a number of basic classes of predications: entities and things
versus relations, and within the relational predicates, stative relations, complex
atemporal relations, and processes. The formal word classes of a language will
typically express a basic type of predication. For instance, while nouns express
the notion of ‘thing’ (a bounded entity in some domain), adjectives will typically
be stative relational predicates.
Second, if you have a grammar with no rules but only symbolic units, how
do you achieve compositionality, i.e. how do you ensure that different symbolic
units may be combined to build larger units, like phrases or sentences? Here,
the trick is to recognize that many predicates have open slots. If, for instance,
the meaning of above is defined in terms of a stative relationship between what
Langacker calls a ‘trajector’ and a ‘landmark’, the trajector and the landmark are
only included schematically, as an open slot, in the meaning of above. Filling
out the slots with other predicates then compositionally yields phrases like the
lamp above the table.
In the course of Langacker’s paper, you will come across a number of concepts
that will play a central role in some of the other chapters included in the present
collection: the schematic network idea (which will come to the fore in Chapter
5), the notion of domain matrix (which will play an essential role in Chapter 8),
and the concept of a construction and a continuum between lexicon and grammar
(which will constitute the focus of Chapter 11 and Chapter 12).
Talmy never suggested a specific label for his approach to grammatical description,
but the label grammatical construal captures very well what he is doing: what are
the forms and patterns of construal (in the sense of conceptual perspectivization
through language) that are realized by the grammatical structure of a language?
This adjective grammatical is quite important here: Talmy focuses on the specific
types of conceptual construal that are expressed by those aspects of natural
language that have to do with syntax and morphology, rather than the lexicon. In
Langacker’s article, we noticed a specific interest in the relationship between the
Introduction: A rough guide to Cognitive Linguistics 9
lexical dimension of the language, and the structural dimension – the syntax and
the morphology. Talmy notes that there are some forms of conceptual construal
that are hardly ever expressed by the grammatical structure (like color), whereas
others (like the distinction between singular and plural) are typically expressed by
syntax and morphology. The bulk of Talmy’s paper, then, provides an overview of
different types of conceptual construal systems that are typical for the structural,
grammatical rather than the lexical subsystem of natural languages.
2.2. The dynamic nature of grammar
If natural language signs are flexible, we will need a model to describe how the
different readings of the expressions relate to each other. Several such models
for the polysemic architecture of expressions have been proposed by Cognitive
Linguistics, and the three concepts in this group describe the most important of
them.
The radial network model describes a category structure in which a central case
of the category radiates towards novel instances: less central category uses are
extended from the center. The paper featured in this collection, ‘Cognitive topology
and lexical networks’ by Claudia Brugman and George Lakoff is based on
Brugman’s seminal analysis of the preposition over. The study was seminal not
just in the sense that it popularized the radial network model, but also because
it spawned a whole literature on the analysis of prepositions (more on this in
the Further reading chapter). Brugman suggests the ‘above and across’ reading
of over (as in the plane flew over) as central, and then shows how less central
readings extend from the central case. These can be concrete extensions, as in a
‘coverage’ reading (the board is over the hole), but also metaphorical ones, as in
temporal uses (over a period of time).
Radial categories constitute but one type of a broader set of models that fall under
the heading of prototype theory. For instance, the importance of specific birds
in the category structure of bird (this is a point we drew the attention to before)
belongs in the same set of phenomena as the radial set idea. The paper ‘Prospects
and problems of prototype theory’ by Dirk Geeraerts presents a systematic
overview of the different prototype-theoretical phenomena that are mentioned in
the literature. Specific attention is paid to the mutual relations that exist among
these phenomena: it is argued that prototype is itself a prototypically structured
10 Dirk Geeraerts
concept, i.e. that there is no single definition that captures all and only the diverse
forms of ‘prototypicality’ that linguists have been talking about.
Prototype theory as described in the previous article is a generalization over
the radial network model. But there is another generalization to introduce: the
schematic network model. What this adds to the radial network and prototype
models is the idea that the dynamism of meaning may also involve a shift along
a taxonomical dimension. This may need some explanation. Note that we can
think of birds at different levels of conceptual abstraction (or schematicity, as it
is also called). At one level, we have a prototypical idea of birds as living beings
that have feathers and wings and that can fly. If we stay on this level, we can move
from the central prototype cases (the ones that correspond to the central concept)
to peripheral cases, like birds without feathers and wings (we mentioned penguins
and kiwis before). But there are other levels at which you can think of birds: more
specific ones (as when you think about individual birds, like your great-uncle’s
parrot) and more general ones (like when you group bird species into categories
like ‘fowl’, ‘birds of prey’, ‘water birds’ etc.).
Moving from a more specific to a more general level is called ‘schematization’,
and the resulting model of readings for an expression is called a schematic
network. The idea of schematic networks is implicit in prototype theory, but it
has been made most explicit by Ronald W. Langacker. The concept plays an
important role in construction grammar (see below), but here, in David Tuggy’s
paper ‘Ambiguity, polysemy and vagueness’, it is applied to a crucial question
about meaning: the relationship between vagueness and polysemy: in a schematic
network, you accept that what is polysemy (different meanings) at one level is
vagueness (a less specified meaning potential) at an other, more schematic level.
In a very clear and graphical way, Tuggy shows how this shift between levels is a
contextual effect: in one situation we may use an expression rather more vaguely,
in another we use it at a more specific, polysemous level.
The consequences for our conception of semantic dynamism are tremendous.
The dynamism of meaning does not just imply that it is easy to add new meanings
to the semantic inventory of an expression, but also that we should not think of
this overall structure of meanings as stable. The semantics of lexical and constructional
units is not a bag of meanings, but is a (prototypically and schematically)
structured meaning potential that is sensitive to contextual effects.
Introduction: A rough guide to Cognitive Linguistics 11
2.3. The non-autonomy of grammar
If meaning is non-autonomous and encyclopedic, it is important to investigate
the way in which different types of experience interact with each other. How
for instance does our bodily experience of a sensory or motor kind relate to our
more abstract thought, and are there any conceptual mechanisms that cut across
the sensorimotor and the abstract mode of human knowledge? The four papers
brought together in this group show how Cognitive Linguistics deals with the
encyclopedic entrenchment of linguistic meaning, from both perspectives: what
is the role of general cognitive mechanisms, and how do specific domains of
experience interact?
Two of the four concepts that will be introduced are not specific for Cognitive
Linguistics: metaphor and metonymy are traditional concepts in natural language
semantics. Cognitive Linguistics has however brought new perspectives to the
study of both metaphor and metonymy, and we will see in what sense. The other
two concepts were newly introduced by Cognitive Linguistics. Incidentally, the fact
that metaphor and metonymy are traditionally known as mechanisms of semantic
change, makes clear that there is a certain degree of overlap between the present
group of concepts and the previous one: some of the concepts mentioned here
may also be seen as illustrations of the dynamic nature of meaning.
Conceptual metaphor is probably the best known aspect of Cognitive Linguistics:
if you’ve heard only vaguely about Cognitive Linguistics, conceptual metaphor
is likely to be the notion that you’ve come across. You will have learnt by now
that there is much more to Cognitive Linguistics, but still, Conceptual Metaphor
Theory occupies a major place in the cognitive linguistic research program. Conceptual
Metaphor Theory rests on the recognition that a given metaphor need not
be restricted to a single lexical item, but may generalize over different expressions.
Such general patterns may then be summarized in an overall statement like
, a pattern that ranges over expressions like the following:
He is known for his many rapid conquests. She fought for him, but his mistress
won out. He fled from her advances. She pursued him relentlessly. He is slowly
gaining ground with her. He won her hand in marriage. He overpowered her. She
is besieged by suitors. He has to fend them off. He enlisted the aid of her friends.
He made an ally of her mother. Theirs is a misalliance if I’ve ever seen one.
This way of thinking about metaphor was introduced in George Lakoff’s and
Mark Johnson’s book Metaphors we live by of 1980, a book that has achieved
something of a bestseller status. The article included here, Lakoff’s ‘The contem-
12 Dirk Geeraerts
porary theory of metaphors’ gives a systematic overview of the theoretical and
practical features of Conceptual Metaphor Theory.
You will notice how the two aspects of non-autonomy that were mentioned
earlier show up in conceptual metaphor theory. First, metaphor is treated as a
general cognitive mechanism, not as a specifically linguistic one that works on
the level of individual expressions. Second, metaphor involves the interaction
between different domains of experience: a source domain (in the example, war)
and a target domain (love). This notion of domain will turn out to be crucially
important for the other concepts in this group as well.
An image schema is a regular pattern that recurs as a source domain (or a structuring
part of a source domain) for different target domains. Typical image schema’s
include containment, path, scales, verticality, and center-periphery. The recurrence
of image schemas may be illustrated by a closer look at the containment
schema. It occurs in conceptual metaphors in which containment is the source
domain for widely diverse target domains like the visual field (in sight, out of
sight, go out of view, inside someone’s field of vision), time (in two hours, he’s
into the first year of his retirement, do something in a short period), difficulties
(get yourself into difficulties, we’re in this together, how do we get out of this, to
be in a mess), obligations (what are you getting into, no way out, can he get out
of it), and the self as contained in the body (withdraw into oneself, a young man
in an old man’s body, there’s an insecure person inside).
Characteristically, image schemas involve some form of sensory or motor
experience, like a spatial configuration in the case of containment. In that sense,
the appearance of image schemas in metaphors is typical for the encyclopedic,
non-autonomous nature of meaning: prelinguistic domains, like the sensorimotor
or spatial ones are mapped onto more abstract domains, providing them with
structure. The notion of image schema, like the notion of conceptual metaphor,
was introduced into Cognitive Linguistics by George Lakoff in his collaboration
with the philosopher Mark Johnson. The paper included here, ‘The cognitive
psychological reality of image schemas and their transformations’ by Ray Gibbs
and Herb Colston, examines the psychological reality of image schemas: how can
you prove, by means of experimental methods, that image schemas do indeed
have a psychological reality?
In this sense, the paper is not just important as an illustration of the notion of
image schema, but also for methodological reasons. The particular theoretical
perspective of Cognitive Linguistics has a number of far-reaching methodological
consequences. For one thing, treating meaning as a mental phenomenon and
focusing on language as a cognitive tool implies that a rapprochement with the
methodology of psychological research is obvious: experimental methods should
Introduction: A rough guide to Cognitive Linguistics 13
bear fruit in Cognitive Linguistics just like they do in psychology. For another,
the idea that grammar is usage-based implies that the analysis of actual usage
data (as in corpora, specifically) should play an important role in Cognitive Linguistics.
The paper by Gibbs and Colston is an example of such an experimental
approach. An example of a corpus-based methodology is presented in Tomasello’s
paper in this volume.
However, neither the use of an experimental method nor the use of corpus
data is as yet a dominant methodology in Cognitive Linguistics. They constitute
emerging tendencies that are likely to gain in importance in the course of the
following years, but a lot of the work done in Cognitive Linguistics is still based
on a more traditional analytic methodology.
There is yet another way in which thinking in terms of domains plays a role in
Cognitive Linguistics, viz. in the analysis of metonymy. In the tradition of lexical
semantics, metaphor and metonymy are distinguished on the basis of the type of
semantic association they involve. Metaphor is supposed to be based on similarity
(if love is war, it is like war in a number of respects), whereas metonymy is said
to be based on contiguity – a somewhat vague notion that could be clarified in
terms of ‘actual proximity or association’. For instance, when you fill up your car,
you don’t fill the entire vehicle with fuel, but only the gas tank. The name of the
whole comes to stand for the part, and part and whole are associated in reality.
Now, metonymy research in Cognitive Linguistics received an important impetus
from the recognition that metonymy could receive a definition that is nicely
complementary to that of metaphor. If metaphor is seen as a mapping from one
domain to the other, metonymy can be seen as a mapping within a single domain.
The shift from whole to part in car is a shift within the physical, spatial domain.
This view on the relationship between metaphor and metonymy was already made
in Lakoff and Johnson’s Metaphors we live by, but the article ‘The role of domains
in the interpretation of metaphors and metonymies’ by William Croft adds an
innovative perspective. The relevant shift, Croft argues, is not necessarily one
within a single domain, but it may be a shift within a domain matrix.
The domain matrix is a notion introduced by Ronald W. Langacker: it captures
the idea that a concept may be simultaneously defined in various domains.
For instance, Shakespeare is not only defined as a physical person, but also in
the literary domain, as an author. So, when you say that you have read the whole
of Shakespeare, you metonymically mean the entirety of his literary production,
rather than the person. What Croft suggests, then, is to define metonymy overall
in terms of such a domain matrix.
14 Dirk Geeraerts
If metaphor is analyzed as a mapping from one domain to another, the question
arises how such mappings take place: how does the structure of the source
domain get mapped onto the target domain? The notion of conceptual integration
developed by Gilles Fauconnier, and represented here by the paper ‘Conceptual
integration networks’ by Gilles Fauconnier and Mark Turner, provides a descriptive
framework to answer that question. It distinguishes between four spaces: a
source input space, a target input space, a blend between both, and a so-called
generic space. For instance (to use an example first described by Seana Coulson),
you can think of ‘trashcan basketball’ as a game in which you throw crumpled
pieces of paper into a trashcan, as you might do in an office environment or in
a student dorm. The game of basketball is one input space, and the office or the
dorm situation the other. The mapping between the two spaces associates the ball
with the piece of paper, the basket with the trashcan, the players with the students
or the office people, and further elaborations are possible. This mapping creates a
blended space, and the relevant features of the blend are not just directly derived
from the original input spaces. On the contrary, you may find emergent structure
that is specific to the blended space: the fact that the trashcan would normally be
placed on the ground, in contrast with a basketball ring, would certainly influence
the way the game is played. The fourth type of space, the generic space, contains
the common structure of the input spaces; in this case, it would be the space of
someone throwing an object into a container.
The description of the four spaces may also explain some of the alternative
names that the conceptual integration approach is known by: the blending or
the mental spaces approach. As mentioned, the conceptual integration approach
clearly links up with the analysis of metaphor as mapping across domains: one
might say that the trashcan example elaborates the metaphor ‘a trashcan is a
basketball ring’. However, the blending analysis is more general than the study
of metaphor. Conceptual integration has proved to be useful in a wide variety of
phenomena, many of which are not even remotely associated with metaphorical
processes. Counterfactuals are a case in point. If Beethoven were alive, he would
use a synthesizer creates a blended space between the present-day musical situation
and the historical space of Beethoven as an innovative composer, but you
cannot really say that the conceptual process is a metaphorical one.
2.4. The experiential grounding of grammar
You will remember that the experiential nature of language raised the question
of the relationship between lexicon and syntax. Cognitive Linguistics provides a
specific answer to that question that links up with what we said about prototype
Introduction: A rough guide to Cognitive Linguistics 15
theory and schematic networks: we can think of a grammar as a schematic network
with abstract patterns at the schematic level, and the lexicalized instantiations
of those patterns (the words and strings of words that fill the patterns) at a more
specific level. From the point of view of mainstream twentieth century linguistics,
this is a bit of a strange idea. If you assume (as generative grammar so vehemently
stressed) that grammar is a set of rules, the lexical items instantiating those roles
are not all that important: you basically need an inventory of items, but the real
work is done by the rules.
However, Ronald W. Langacker has pointed out that there is a fallacy here: the
so-called rule/list fallacy, i.e. the idea that what can be handled by rules should
not be listed. If you start from that assumption, you get a strict separation between
lexicon and syntax. But if you think in terms of schematic networks, such a separation
is not necessary at all: you can both describe abstract patterns and their
concrete lexicalizations. In fact, you have to, because there are prototypicality
effects that you may need to describe. An indirect object construction of the type
Subject – Verb – Direct Object – Indirect Object is typically filled by verbs like
give or tell, and less typically so by verbs like envy. It is part of our grammatical
knowledge that we recognize those typicalities (just like we recognize blackbirds
as more typical birds than ostriches).
The specific form in which this idea is realized in Cognitive Linguistics, is
in the form of construction grammars. The plural is deliberate here: since there
are various forms of construction grammar, Construction Grammar is a family
of theories rather than a single well-defined approach. The three papers in this
group all involve Construction Grammar, but it is the paper by Goldberg (Chapter
11, the second one in the group) that introduces the approach most directly. The
first paper in the group presents frame semantics, which is one of the important
sources for Construction Grammar. The final paper show how the principles of
Construction Grammar can be applied in language acquisition research.
Frame semantics is the specific approach to natural language semantics developed
by Charles Fillmore. The article included here, ‘Frame semantics’, sets out
the basics of the theory. One essential starting-point is the idea that one cannot
understand the meaning of a word (or a linguistic expression in general) without
access to all the encyclopedic knowledge that relates to that word. This obviously
ties in with the non-autonomous nature of natural language semantics: that meaning
in natural language is not separated from other forms of knowledge implies
that it is not very useful to maintain a strict separation between world knowledge
and knowledge of linguistic meaning. While this recognition would be shared
by all forms of semantics in Cognitive Linguistics, the individual identity of
16 Dirk Geeraerts
frame semantics involves the specific structures of encyclopedic knowledge that
it invokes. Basically, these ‘frames’ are things happening and occurring together
in reality. For example, in order to understand the word sell, you need to have
world knowledge about the situation of commercial transfer. This comprises, apart
from the act of selling, a person who sells, a person who buys, goods to be sold,
money or another form of payment, and so on.
A semantic frame of this type is a coherent structure of related concepts where
the relations have to do with the way the concepts co-occur in real world situations.
Knowledge of the frame is necessary for an adequate knowledge of the
words referring to the concepts in the frame: a word activates the frame, highlights
individual concepts within the frame, and often determines a certain perspective
in which the frame is viewed. In the standard commercial transaction example,
for instance, sell construes the situation from the perspective of the seller and buy
from the perspective of the buyer.
Although frame semantics was originally applied predominantly to the semantic
description of words, there is a close relationship with Construction Grammar.
Words like sell come with their own set of constructions (like sell something to
someone or sell something for a certain price), and the different constructions
reflect different ways in which the frame can be highlighted. In this way, frame
semantics can be integrated with Construction Grammar as one way of specifying
the semantics of constructions. Overall, we can now see that frame semantics
occupies a transitional position in our grouping of concepts. On the one hand, if
we focus on the way in which it uses structured encyclopedic knowledge as the
background for the description of meanings in natural language, it belongs with
the previous group of papers: it describes one of the ways in which conceptual
knowledge of an encyclopedic (i.e. not specifically linguistic) nature is structured.
On the other hand, if we concentrate on the input that frame semantics provides
for the description of construction types, it links up with the present group of
articles.
Simplistically, a grammatical construction is any string of words or morphemes
exhibiting a coherent pattern, whether it be an entire sentence or a clause or a
phrase (like a noun phrase or a verb phrase) or a complex lexeme (like a phrasal
verb). And of course, the abstract pattern itself may also be called a ‘construction’.
In classical contemporary grammars, an indirect object pattern of the type Subject
– Predicate – Indirect Object – Direct Object would be considered a derived
structure, built up from the functional classes Subject, Predicate etc. That is to
say, the rules of the grammar would be defined in such a way that a grammatical
pattern of the form Subject – Predicate – Indirect Object – Direct Object could
Introduction: A rough guide to Cognitive Linguistics 17
be assembled on the basis of the relevant functional building blocks. In Cognitive
Linguistics, a pattern of this type is considered to be non-derived, i.e. is taken to be
a sign of the language, or, if you wish, an (abstract) expression in its own right.
As you can read in Adele Goldberg’s paper ‘The inherent semantics of argument
structure: The case of the English ditransitive construction’, there are a number
of advantages to such an approach, two of which may be mentioned here, to give
you a first idea of why thinking in terms of constructions may be interesting.
First, if a construction is treated as an entity in its own right, it is possible that
the whole has characteristics that cannot be straightforwardly derived from the
constituent components. This is a property known as non-compositionality: the
meaning of the whole is not necessarily a compositional function of the meaning
of the parts. Second, if any construction is a distinct element of the inventory of
linguistic signs for a given language, you can treat constructions like any other
category – according to the usual practice of Cognitive Linguistics. That is to say,
it will then be quite normal to describe not just the meaning and the form of the
category, but also its salient members. In the case of constructions, we may then
think primarily of the lexical elements that can fill the slots of the construction
(like the verbs that occupy the predicate role in the indirect object construction).
Or in other words: describing grammar as a schematic network with lexical elements
at the bottom and more abstract patterns higher up in the network is completely
congenial to a construction grammar approach.
Defining Cognitive Linguistics as a usage-based model of language has a number
of consequences, like the straightforward methodological conclusion that cognitive
linguists will have to invest in the analysis of real language use if they are to
live up to their self-declared status. You would expect a lot of corpus research in
Cognitive Linguistics, then, but to be honest, this is a trend that is clearly emerging,
but that has not yet gained as prominent a status as one would expect.
Using corpora of observed speech is natural in language acquisition research,
though: if you wish to study how children acquire their language, you will obviously
want to observe and analyze their developing language. Also, language
acquisition is a domain par excellence to test a usage-based model of language
that believes that our experience of actual speech determines how we come by
more abstract patterns. In fact, if our more schematic knowledge is based on our
more concrete knowledge of lexical instantiations of such patterns, you should
be able to observe ‘lexical bootstrapping’ effects: the specific words in which we
begin to learn certain constructions, determine how we learn the construction.
That is exactly what Michael Tomasello investigates in his paper ‘First Steps
toward a Usage-based Theory of Language Acquisition’, which is part of a long-
18 Dirk Geeraerts
term research effort to develop a theory of language acquisition that ties in with
Cognitive Linguistics and Construction Grammar.
If you are familiar with the history of contemporary linguistics, you will
appreciate how important such an attempt is. The generative grammar idea that
language is a separate module of the mind is very much based on an argument
from language acquisition: if we do not assume that language is genetically wired
in, the argument goes, we couldn’t explain at all how the acquisition of language
proceeds so quickly as it actually does. This is particularly the case, the argument
continues, because the input children get (the language they are exposed
to) is not sufficient to explain how they could learn all the intricacies of natural
language syntax. This is the ‘poverty of stimulus’ argument. Obviously, if Tomasello
succeeds in his attempt to explain how children can learn language through
abstraction from the actual input they get, a central tenet of generative grammar
will be overturned.
2.5. A conceptual map
Let us summarize. What we have shown in this (somewhat detailed) introduction
to the tour of twelve papers is how they introduce, discuss, illustrate concepts
that follow logically from the central characteristics of Cognitive Linguistics
that we learned about in the previous section. It will not come as a surprise to
you that these twelve concepts are far from being the only relevant or interesting
ones that have been developed in the context of Cognitive Linguistics – but they
are certainly among the most basic ones. When you read the individual papers,
it may be wise to refer back once in a while to the characterizations that you find
in the foregoing pages: they will help you not to lose track and to interpret the
conspicuous features of the papers.
In Figure 1, the relations between the various papers are graphically represented.
The fundamental features of Cognitive Linguistics (the perspectival,
dynamic, non-autonomous, experiential nature of natural language) are mentioned
in the corners of the figure. Intermediate between these cornerstones and the
twelve central concepts, you will find six fields of research that emerged in the
foregoing discussion as the logical link between the fundamental features and the
central concepts: the conceptual characterization of the grammar, the search for
models of polysemy, the analysis of mechanisms of polysemy, the importance of
thinking in terms of domains of experience, the relationship between grammar
and lexicon, and the relationship between structure and usage. The arrows in the
figure indicate how each more specific concept is motivated by a more general one
for instance: looking at mechanisms of categorial polysemy is one way of getting
a grip on the dynamic nature of linguistic meaning, and in a next step, conceptual
metaphor is one of the specific mechanisms of categorial polysemy.
Introduction: A rough guide to Cognitive Linguistics 19
Figure 1. A conceptual map of Cognitive Linguistics
20 Dirk Geeraerts
Adding the intermediate level with the six fields of research brings in extra subtlety
in the overview. We can now make clear, for instance, that Langacker’s paper
with which the collection opens, not only illustrates the perspectival nature of
grammar, but also deals with the relationship between lexical structure and morpho-syntax.
In the same way, the figure illustrates how the experiential nature
of the grammar does not only involve questions about the acquisition and the
architecture of the grammar, but also links up with thinking in terms of domains
and their mutual relations.
3. Where do you go next?
Let us assume that, after roaming the present introductory volume, you really
like the look and feel of Cognitive Linguistics. It’s a safe assumption, in fact: you
are bound to be drawn in by an intellectual climate that is both hospitable and
inspiring, open-minded and exciting, wide-ranging and innovative. But where do
you go after the initial tour d’horizon that has won your heart? Let’s go over a
few trajectories that might cater to your personal interests.
A first thing to do would be to complete your initiation by reading the companion
volume to the present collection. In several ways, Cognitive Linguistics:
Current Applications and Future Perspectives (edited by Gitte Kristiansen, Michel
Achard, René Dirven and Francisco J. Ruiz de Mendoza) is deliberately complementary
to the book that you are holding in your hands: it does not focus on Cognitive
Linguistics basics, but rather describes the contemporary state of affairs in
the main fields of application of Cognitive Linguistics. Also, it does not consist
of existing papers, but only contains newly written articles that have the explicit
purpose of presenting current discussions and domains. Taken together, the two
volumes will thoroughly familiarize you with Cognitive Linguistics, way beyond
the fragmented and uncertain knowledge that an incidental visit would impart.
From that point on, you may want to go beyond the level of introductions,
and you may, with the acquired confidence of the experienced traveler, do some
journeying on your own. However, in case your self-assurance has not reached an
optimal level and you would want to boost it with an additional round of introductory
reading, there are some good book-length introductions to Cognitive
Linguistics that may help you. Here’s a list of commendable texts.
Friedrich Ungerer and Hans-Jörg Schmid. 1996. An Introduction to Cognitive
Linguistics. London/New York: Longman.
Violi, Patrizia. 2001. Meaning and Experience. Bloomington: Indiana University
Press.
John R. Taylor. 2003. Linguistic Categorization. Third Edition. Oxford:
Oxford University Press.
•
•
•
Introduction: A rough guide to Cognitive Linguistics 21
René Dirven and Marjolijn Verspoor. 2004. Cognitive Exploration of Language
and Linguistics. Second Revised Edition. Amsterdam, Philadelphia:
John Benjamins.
William Croft and D. Alan Cruse. 2004. Cognitive Linguistics. Cambridge:
Cambridge University Press.
Vyvian Evans and Melanie Green. 2006. Cognitive Linguistics. An Introduction.
Edinburgh: Edinburgh University Press.
Once you are ready to leave the introductory level, there are basically two things
to do: deepen your understanding of the existing body of work in Cognitive Linguistics,
and keep aware of current developments. With regard to the former, a
first set of routes to explore is provided by the present volume itself: accompanying
each of the chapters included in the collection, you find a broadly conceived
set of suggestions for further reading in the Epilogue. They will direct you to
elaborations and discussions of the basic concepts that are presented here – and
to some of the less basic (but no less important) concepts developed in the field
of Cognitive Linguistics. Alternatively (or in parallel), you could have a look at
the forthcoming Handbook of Cognitive Linguistics, edited by Dirk Geeraerts
and Hubert Cuyckens and published by Oxford University Press. It’s a collection
of some fifty commissioned articles that each offer an in-depth treatment of one
of the manifold aspects of Cognitive Linguistics.
To keep in touch with new work, you would certainly want to be aware of
new publications. So, what are the journals and the book series that you need to
keep an eye on? Note that a lot of studies in Cognitive Linguistics are now being
published in general publication channels and by a wide variety of international
publishers. Here, only journals and book series that are specifically dedicated to
Cognitive Linguistics are mentioned.
Two journals need to be cited: Cognitive Linguistics, and the Annual Review
of Cognitive Linguistics. The former, published by Mouton De Gruyter, is the
official journal of the International Cognitive Linguistics Association (more about
which in the next section). The journal was founded by Dirk Geeraerts in 1990.
Consecutive editors-in-chief so far were Arie Verhagen and Adele Goldberg.
The latter journal, published by the John Benjamins Publishing Company under
the auspices of the Spanish Cognitive Linguistics Association, first appeared in
2003. It is led by Francisco Ruiz de Mendoza. The journal Cognitive Linguistics
is not only the most reputable journal in the field, it also comes with a considerable
bonus. A subscription to Cognitive Linguistics includes a copy of the digital
Bibliography of Cognitive Linguistics – and indispensable bibliographical tool
compiled through the relentless efforts René Dirven (undoubtedly the major
organizational force behind the entire Cognitive Linguistics enterprise). The
bibliography now covers 7000 publications (no, you won’t have to read them all
•
•
•
22 Dirk Geeraerts
to be recognized as an accomplished cognitive linguist), and it will be regularly
updated in the following years.
Five book series specifically dedicated to work in Cognitive Linguistics need
to be mentioned. Cognitive Linguistic Research or CLR is the oldest and most
complete series. Published by Mouton de Gruyter of Berlin, it was launched at the
same time as the journal Cognitive Linguistics, and now numbers over thirty volumes.
Recently, Cognitive Linguistic Research (with Dirk Geeraerts as managing
editor) has received a companion series in the form of Applications of Cognitive
Linguistics or ACL (managed by Gitte Kristiansen) – a series that will focus on
the descriptive applications of Cognitive Linguistics, while Cognitive Linguistic
Research harbors the more theoretically relevant studies. The other main publisher
for Cognitive Linguistics, the John Benjamins Publishing Company of Amsterdam,
features three specifically relevant (but as yet less extensive) series. While
Human Cognitive Processing (edited by Marcelo Dascal, Raymond W. Gibbs
and Jan Nuyts) has a broad cognitive orientation, the two other series are more
specific: Cognitive Linguistics in Practice (edited by Günter Radden) covers the
field from a textbook oriented perspective, and Constructional Approaches to
Language (edited by Mirjam Fried and Jan-Ola Östman) specifically deals with
Construction Grammar.
If you are interested in following what is going on in Cognitive Linguistics,
note that the book series are not just monograph series. A lot of what is being
published in the series consists of collective volumes with thematically interconnected
articles. In many cases, these constitute selections of papers presented at
Cognitive Linguistics conferences. Because quite a lot of relevant work is being
published in such collective volumes, you will profit from keeping an eye on them
once you’ve become a Cognitive Linguistics aficionado.
When you’ve reached this stage, you will be ready to take a step into the world
and take part in some real life Cognitive Linguistics activities. Where would you
go? All self-respecting cities and countries have their own festivals and fiesta,
and becoming part of the crowd involves participating in the celebrations. In
Cognitive Linguistics, the main community event is without any doubt the biannual
ICLC or Cognitive Linguistics Conference. The first ICLC took place in
1989 in Duisburg, Germany. It was one of the so-called LAUD symposia (where
LAUD stands for Linguistic Agency of the University of Duisburg) that had been
organized by René Dirven since 1977 and where some of the world’s most distinguished
linguists were invited to present their work. The Duisburg conference was
of crucial importance for the institutionalization and the international expansion
of Cognitive Linguistics: it was there and then that the International Cognitive
Linguistics Association or ICLA was founded (the conference was accordingly
rebaptized as the First International Cognitive Linguistics Conference), that plans
were made to launch the journal Cognitive Linguistics, and that the monograph
Introduction: A rough guide to Cognitive Linguistics 23
series Cognitive Linguistics Research was announced. Cognitive Linguistics as
an intellectual movement is too self-critical to recognize any historical sites or
places of pilgrimage, but if ever a commemorative plate were to be considered,
Duisburg would be a likely candidate.
The following ICLC conferences were consecutively held in Santa Cruz, US
(1991, organized by Gene Casad), Leuven, Belgium (1993, organized by Dirk
Geeraerts), Albuquerque, US (1995, organized by Sherman Wilcox), Amsterdam,
The Netherlands (1997, organized by Theo Jansen and Gisela Redeker), Stockholm,
Sweden (1999, organized by Erling Wande), Santa Barbara, US (2001, organized
by Ronald W. Langacker), Logroño, Spain (2003, organized by Francisco J. Ruiz
De Mendoza), and Seoul, South Korea (2005, organized by Hyon-Sook Shin).
The tenth ICLC in 2007, organized by Elzbieta Tabakowska, has its venue at the
Jagiellonian University of Krakow in Poland.
Next to the ICLC’s, there are a number of local events that you may consider
attending. The LAUD conferences are now being organized bi-annually in Landau,
Germany, in the year between the ICLC conferences. That holds for a number
of other regular meetings, like the US-based Cognitive Structure, Discourse and
Language conference, which reached its eighth edition in 2006; or the International
Conference on Construction Grammar, which had its fourth edition in 2006.
An important part in the organization of workshops and symposia in Cognitive
Linguistics is currently played by the ICLA affiliates. These are ICLA branches
defined by region or country (and occasionally by language). The first one to be
founded was the Spanish Cognitive Linguistics Association, whose affiliation
was formally approved at the 1999 ICLC. The year 2001 saw the affiliation of
the Finnish, the Polish, and the Slavic Cognitive Linguistics Associations. Further
affiliates include the Russian Association of Cognitive Linguists (2004), the
German Cognitive Linguistics Association (2005), the Discourse and Cognitive
Linguistics Association of Korea (2005), the Association Française de Linguistique
Cognitive (2005), the Japanese Cognitive Linguistics Association (2005),
the Conceptual Structure, Discourse and Language Association (2005), and the
UK Cognitive Linguistics Association (2006). Further local branches are likely
to emerge, and all of them are likely to organize regular meetings. If you are
interested in following what is on the agenda, you may want to consult the ICLA
website: it contains a calendar of Cognitive Linguistics events, and supplies links
to the individual websites of the ICLA affiliates. This is the ICLA address: http://
www.cognitivelinguistics.org/
Meetings, lectures, workshops, symposia, and conferences are also announced
on Cogling, a mailing list for disseminating ICLA news, queries, and discussions
of interest to cognitive linguists. Again, details about subscribing may be
found on the ICLA website. Incidentally, the website also offers an overview of
24 Dirk Geeraerts
the courses or programs in Cognitive Linguistics that are offered at many places
around the globe.
Once you get to one of the conferences, who would you be likely and/or eager
to meet? Thinking in terms of people, the key figures of Cognitive Linguistics
are George Lakoff, Ronald W. Langacker, and Leonard Talmy. Round this core
of founding fathers, who originated Cognitive Linguistics in the late 1970s and
the early 1980s, two chronologically widening circles of cognitive linguists may
be discerned. (The lists that follow are obviously indicative only: they are not
meant to exclude anyone, but only to give you an idea of the different ‘generations’
of cognitive linguists.)
A firstwave, comingto the fore inthe secondhalfofthe 1980s and the beginning
of the 1990s, consists of the early collaborators and colleagues of the key figures,
together with a first generation of students. Names that come to mind include those
of Gilles Fauconnier, Eve Sweetser, Mark Johnson, Mark Turner, Raymond W.
Gibbs, William Croft, Adele Goldberg, Dave Tuggy, Gene Casad, Laura Janda,
SuzanneKemmer,SallyRice,RicardoMaldonado,KarenVanHoek,GeoffNathan,
Margaret Winters, Sherman and Phyllis Wilcox, Margaret Freeman.
Simultaneously, a number of people in mostly Western and Central Europe took
up the ideas of Cognitive Linguistics and contributed to their international dissemination.
Names include those of René Dirven (to repeat: his instrumental role in the
expansion of Cognitive Linguistics can hardly be overestimated), Brygida RudzkaOstyn,
John Taylor, Zoltan Kövecses, Chris Sinha, Brigitte Nerlich, Arie Verhagen,
Barbara Lewandowska-Tomaszczyk, Elzbieta Tabakowska, Peter Harder, Günter
Radden, Susanne Niemeier, Martin Pütz, Hans-Jörg Schmid, Hubert Cuyckens
and the author of the present introduction.
The mid 1990s and later witnessed a second wave of expansion, with second
generation students and a further geographical spread directed largely towards Asia
and the rest of Europe. Names include those of Alan Cienki, Michel Achard, Joe
Grady, Tim Rohrer, Seana Coulson, Todd Oakley, Gary Palmer, Jose M. GarciaMiguel,
Antonio Barcelona, Francisco Ruiz de Mendoza, Carlos Inchaurralde,
Andrej Kibrik, Ekaterina Rakhilina, Michael Tomasello, Ted Sanders, Wilbert
Spooren, Gerard Steen, Stefan Grondelaers, Stefan Gries, Anatol Stefanowitsch,
Yo Matsumoto.
In addition, you might profit from the occasion to rub elbows with people
who would perhaps not describe themselves unreservedly as cognitive linguists
(coming as they do from other theoretical families or other disciplines, or simply
because they like their independence), but who would show up at the Cognitive
Linguistics conferences because they have relevant things to say: linguists like
Charles Fillmore, Joan Bybee, Elizabeth Traugott, Östen Dahl, Jan Nuyts, or
psychologists like Melissa Bowerman, Dedre Gentner, and Dan Slobin.
Introduction: A rough guide to Cognitive Linguistics 25
4. Why would you want to come back?
So now you know your way around in Cognitive Linguistics. You can walk the
walk and talk the talk, and there’s no way that you’d be exposed as a novice. But
why would you be coming back? What would be a good reason to become a permanent
resident? An obvious but relatively superficial motivation would be the
diversity of the panorama: there’s a lot to be found in the Cognitive Linguistics
archipelago, and the framework is not so strict as to stifle creativity. It’s a lively,
colorful, varied environment, and you’re likely to find some corner of special
significance to you, where you can do your thing and meet people with similar
interests. But beyond that? What would be the long-term importance of Cognitive
Linguistics?
Let us try to take a bird’s eye view of the history of linguistics, and see exactly
where Cognitive Linguistics fits in, and why it could be important for the future
of linguistics. Agreed, you can only achieve this sort of extreme synthesis if you
allow for massive simplification: let us try to keep that in mind as a proviso when
we do the exercise.
The development of linguistics in the twentieth century, then, is characterized
by a succession of two dominant approaches: the structuralist one, and the
generativist one. Currently, in the first decade of the 21st century, the generativist
paradigm is no longer the principal framework, but there clearly is no new central
approach yet. If one looks at Cognitive Linguistics from this perspective, there are
indications that Cognitive Linguistics combines a number of tendencies that may
also be found in other contemporary developments in theoretical linguistics – viz.
in the broad range of functionalist approaches to linguistics. By combining these
tendencies, Cognitive Linguistics taps directly into the undercurrent of contemporary
developments. Specifically, if we recognize that decontextualization is a
fundamental underlying characteristic of the development of grammatical theory
in twentieth century linguistics, Cognitive Linguistics strongly embodies the
recontextualizing tendency that is shared by most functionalist approaches.
The logic behind the decontextualization of twentieth century grammar may
be grasped if we take our starting-point in De Saussure, the founding father of the
structuralist approach. The Saussurean distinction between langue (the language
system) and parole (the use of the language system in actual usage) creates an
internally divided grammar, a conception of language with, so to speak, a hole
in the middle. On the one hand, langue is defined by De Saussure as a social
system, a set of collective conventions, a common code shared by a community.
On the other hand, parole is an individual, psychological activity that consists of
producing specific combinations from the elements that are present in the code.
When langue and parole are defined in this way, there is a gap between both: what
is the factor that bridges the distance between the social and the psychological,
26 Dirk Geeraerts
between the community and the individual, between the system and the application
of the system, between the code and the actual use of the code?
The father of generative grammar, Noam Chomsky, provided an answer by
introducing a distinction between competence and performance: the missing link
between social code and individual usage is the individual’s knowledge of the
code. Performance, in the Chomskyan sense, is basically equivalent with Saussurean
parole, but competence interiorizes the notion of the linguistic system.
Competence is the internal grammar of the language user, the knowledge that
the language user has of the linguistic system and that he or she puts to use in
actual performance.
Remarkably, however, Chomsky introduces a new gap into the system. Rather
than the threefold classification that one might expect, he restricts his conception
of language to a new dichotomy: the social aspects of language are largely ignored.
In comparison with a ternary distinction distinguishing between langue, competence,
and parole/performance (between social system, individual knowledge
of the system, and individual use of the system), the binary distinction between
competence and performance creates a new empty slot, leaving the social aspects
of language largely out of sight.
This apparent lack of interest for language as a social sign system links up
logically with the Chomskyan emphasis on the genetic nature of natural language.
Where, in fact, does the individual knowledge of the language come from? If the
source of linguistic knowledge is not social, what else can it be than an innate and
universal endowment? If the language is not learned through acculturation in a
linguistic community, what other source could there be for linguistic knowledge
except genetics?
Further restrictions follow. Meanings constitute the variable, contextual, cultural
aspects of language par excellence. Because social interaction, the exchange
of ideas, changing conceptions of the world are primarily reflected in the meaning
of linguistic expressions, meanings are less interesting from a genetic point of
view. Also, if the lexicon is the main repository of linguistically encoded meaning,
studying the lexicon is of secondary importance: the focus will fall on the abstract
syntactic patterns. And finally, if linguistics focuses on formal rule systems, the
application of the rule systems in actual usage is relatively uninteresting. If the
rules define the grammar, it is hard to see what added value could be derived from
studying the way in which the rules are actually put to use. The study of performance,
in other words, is just as secondary as research into the lexicon.
In short, generative grammar led to a severe decontextualization of the grammar,
separating the autonomous grammatical module from different forms of context:
through the basic Chomskyan shift from langue to competence, linguistics is
separated from the social context of language as a social code; through the focus
on the genetic aspects of the language, linguistics is separated from the cognitive
Introduction: A rough guide to Cognitive Linguistics 27
context of lived individual experience; through the focus on formal rule systems,
linguistics is separated from the situational context of actual language use.
Compared to this, Cognitive Linguistics is very much a recontextualizing
approach. First, it is an outspoken attempt to give meaning a central position in
the architecture of the grammar. Second, in contrast with formal semantics, the
conception of meaning that lies at the basis of this approach is not restricted to
a referential, truth-functional type of meaning – the type of meaning that you
could express in logical terms. Linguistic structures are thought to express conceptualizations,
and conceptualization goes further than mere reference. As we
have seen, it involves imagery in the broadest sense of the word: ways of making
sense, of imposing meaning. Also, the conceptualizations that are expressed
in natural language have an experiential basis, i.e., they link up with the way in
which human beings experience reality, both culturally and physiologically. In
this sense, Cognitive Linguistics embodies a fully contextualized conception of
meaning. Third, the link between linguistic performance and grammar is reestablished
by the view that language is usage-based, i.e. that there is a dialectic
relationship between langue and parole.
We can observe, then, that the various characteristics of Cognitive Linguistics
that you learned about in the first section of this introductory chapter and that
were further spelled out in twelve crucial concepts, can be summarized under
one general denominator: the recontextualization of grammar. If we assume,
next, that this recontextualizing tendency is an underlying trend in contemporary
cognitive-functionalist linguistics, Cognitive Linguistics is probably one of the
most outspoken representatives of this tendency. Surely, it is not the only one,
and there are whole disciplines in linguistics that are devoted to the exploration
of specific forms of context, like sociolinguistics dealing with the social context,
or pragmatics and text linguistics dealing with the level of parole. But Cognitive
Linguistics is specific in the extent to which it tries to integrate these different
tendencies into an overall model of language.
All of this does not mean, however, that Cognitive Linguistics has reached its
goal yet. To begin with, because it is far from being alone in pursuing a recontextualized
line of linguistic research, one of the major tasks for its future development
will be to systematically confront similar approaches within the broad cognitivefunctionalist
domain and see to what extent theoretical and empirical convergences
are possible. That is not going to be self-evident, given that Cognitive Linguistics
is not even a theoretically uniform framework on its own account.
Further, we cannot say that Cognitive Linguistics has already realized all the
consequences of the decontextualizing stance. For one thing, seeing language in the
context of the mind at large would seem to imply a lively interdisciplinary interaction
with the other sciences of the mind, but that is a trend that is only gradually
emerging in mainstream cognitive linguistic circles. For an other, recontextual-
28 Dirk Geeraerts
izing language in its social context implies an awareness of the variation that is
inherent in the social life of language. Here again, sociolinguistically oriented
studies inspired by Cognitive Linguistics are only beginning to come into view.
In short, there is still a quite a lot to be done. Ultimately, that may well be
the best reason for coming back: not what has already been achieved, but what
still has to be done – all the exciting, inviting paths for further exploration. So
we will meet there, right?