The Architecture of a „Grammar of Scenes and Scenarios”
Wolfgang Schulze (University of Munich)
[Paper of talk given at the TABU day (U Groningen 1999), version corrected by Joseph Swango]
[Numbers in the text refer to the hand out (RTF format)]
Contents:
1. Typology and Cognitivism
2. The Grammar of Scenes and Scenarios
3. Some Conclusions
The history of linguistic research can easily be described with certain
basic scientific paradigms that have dominated the objectives, methods,
and empirics of this discipline since at least the age of humanism. These
paradigms oscillate between two poles that were titled philosophical and
harmonic traditions. In modern terms we can tentatively equate the philosophical
tradition to deductive approaches, whereas harmonic paradigms tend towards
inductive methods. In the heydays of philosophical grammars, that is between
say 1750 and 1850, it was rationalism, especially its variant in terms
of Wolffian logicism that dominated the explanatory access to language.
Today, this explanatory basis has become much broader and less restricted
to a single causal paradigm. Though the mental domain still plays the most
important role in this respect. Other parameters such as communication,
cultural and social knowledge, the habitus of a speech community, as well
as parameters anchored in the interaction of human biology and cognition
have partly led to the assumption of a multicausal scenario in order to
access the ontology of language.
The harmonic paradigm mainly is an output of early
European romanticism based on the tradition of sensualism and French encyclopedias.
It yields in the description of the grammatica vulgaris or grammaire particulière
of a language in order to provide the basis for establishing a more general
typology often based on a naturalist interpretation of verbal behavior.
Though comparisons have been introduced as a descriptive method already
in classical times that have also been exploited by the tradition of the
Grammar of Port Royal in the 17th century. It was the non-rationalistic
movement of early Romanticism by which this method became popular. It gradually
undermined the philosophical tradition also because colonialism provided
the Europeans with a growing knowledge of exotic languages that questioned
the structures of which most of the early axioms related to the ontology
of language were built upon.
From a historical perspective we can describe a
significant interaction of methodical paradigms and the quantity of linguistic
materials (cf. (1)): If the amount of data processed during a period of
inductive research reaches a critical mass the paradigm may face a drastic
shift to deductive procedures. (1) Indicates such periods of inductivism
with the help of the label Mithridates. By this term I refer to the two
Mithridates publications listed in (2). Both of Gesner’s Mithridates, de
differentiis linguarum from 1555 and Adelung’s Mithridates or Allgemeine
Sprachenkunde from 1806-1817 represent typical efforts to document the
contemporary encyclopedic knowledge of linguistic diversity. Until now
we have to assume that at least four such Mithridates periods, the last
of which mainly started in two centers after World War II. The first one
was located in Paris in 1945. A year that is associated with the beginning
of a series of meetings organized by the Société de Linguistique
de Paris and devoted to problems of mass comparison. Second, in 1951 a
first informal meeting of linguists, anthropologists, and psycholinguists
took place at Cornell University that represented the first steps towards
the famous Dobbs Ferry Conference of 1961. This conference prepared the
ground for an increasing interest in language typology and language universals
described from a radical inductive point of view. Contrary to earlier Mithridates
periods, the inductive tradition survived despite of the challenge that
emerged from the growing MIT paradigm the propagation of which started
especially with Chomsky’s review of Skinner’s Verbal Behavior in 1959.
Since then, inductive and deductive procedures of
linguistic argumentation have coexisted in a more or less pronounced form
of apartheid.Linguistic research is normally classified according to such
labels as typology or syntax which represent the two sides of the linguistic
war that still takes place between East Coast and West Coast Linguistics.
Whereas the MIT paradigm strongly maintained a position
that followed the basic axioms of ineptness and Universal Grammar. Typology
underwent several substantial shifts in the last four decades that also
documented a shift from inductive to more deductive procedures. This shift
may be explained by the dynamics of the forth Mithridates period itself,
but obviously it also was the academic success of syntax theories that
promoted a gradual movement of the typological paradigm towards some methodical
assumptions often related to the MIT orthodoxy. The two terms that best
document this paradigmatic shift are universals and cognition.
If we try to classify the changes that have taken
place in the theoretical and methodical paradigms of typology it seems
useful to refer to Croft’s terminology (Croft 1990). He distinguishes between
typological classification, typological generalization, and typological
approach or typology proper. These labels correspond to three distinct
types of typological tradition that started in the beginning of the 19th
century, cf. (3). The strongly inductive approach of typological classification
can be characterized as a pre-theoretical way to account for the diversity
of language systems. It had its heydays in the last century, but still
formed a part of the paradigm of Classical American Structuralism. The
above mentioned Dobbs Ferry Conference opened the way to develop a methodological
basis for comparative research in terms of linguistic universals. It is
interesting to note that the Dobbs Ferry Conference, sponsored by the Committee
on Linguistics and Psychology of the Social Science Research Council already
anticipated the interdisciplinary orientation that is characteristic for
the third phase in typological research. Namely Typology proper: Among
the members of the preparatory meeting at Bloomington in 1953 (cf. 4) we
find six linguists, six psychologists, and one anthropologist. The Hullian
psychologist Charles E. Osgood commented upon the meeting as follows (cf.
5): (...) while linguists had an admirable and well worked out method,
it was applied merely to the description of individual languages. Could
the linguists present tell him anything about all languages? That would
be of the highest interest to psychologists” (Greenberg et al. 1978:v).
The famous Memorandum Concerning Language Universals, formulated by Joseph
H. Greenberg, Charles E. Osgood, and James J. Jenkins and circulated among
the participants of Dobbs Ferry in 1961, stressed the interdisciplinary
option (cf. (6)): „(...) since language is at once both an aspect of individual
behavior and an aspect of human culture, its universals provide the major
point of contact with underlying psychological principles (psycholin-guistics)
and the major source of implications for human culture in general (ethnolinguistics)”
(Greenberg et al. 1978:xv).
Yet, it took another 20 years before parts of the
community of typologists adopted the scientific implications underlying
the Memorandum. This step can be characterized as a shift towards a more
explanatory paradigm. Naturally, the orientation of this paradigm was co-determinated
by the cognitive turn associated with the MIT orthodoxy. Some of its renegades
such as George Lakoff, James McCawley, and the psychologist George
Miller entered the paradigm of functionalism and typology and heavily influenced
the way explanatory domains were established for linguistic phenomena.
These anchor domains can be identified among others with the help of the
term’s communication, interaction, culture, habitus, cybernetics, and cognition.
Today in fact, communication and cognition represent the two most basic
domains which, however, are often though to stand in a somewhat antagonistic
relationship, cf. (7): The scheme also indicates that features such as
culture, habitus, and interaction represent secondary domains that can
be claimed by both the communicative and the cognitive paradigm. In more
recent years, the diachronic perspective as a device to explain linguistic
data has gained special interest both with respect to system immanence
and the dynamics of the anchor domains themselves.
The impact of MIT cognitivism on typological research
can hardly be described in terms of a well formulated theoretical debate.
Rather, this impact has to be characterized as implicit and tacit. Rather,
it was the output of what is often called the Roschian Revolution, that
is the psychological research in the tradition of Eleonor Rosch, that was
often pronounced as the framework underlying cognitive thinking in typology.
The LAUD symposium held in Duisburg in 1990 marked „the birth of cognitive
linguistics as a broadly grounded, self-conscious intellectual movement”,
as Langacker has put it.
Although non-MIT oriented cognitive linguistics
can certainly be characterized as the most rapidly expanding linguistic
paradigm of the last decade, the impact of this linguistic theory on standard
work in typology still lacks an adequate interpretation.
Today, Cognitive Linguistics – as opposed to Chomskyan
Cognitivism– are interested especially in the conceptual organization of
linguistic knowledge, cf. (8) which lists parts of the issues that are
intended to be discussed at the forthcoming LAUD symposium in Konstanz
in 2000:
· linguistic categories and cognitive models
· conceptual metaphors and metonymies, e.g. emotion concepts
· schemas and prototypes exploited and built up in and across
language(s)
· construals within languages and across languages
· iconicity in language and thought
· language and space as the basis of concrete and abstract conceptualization
· language, culture and thought, i.e. language as a culture’s
symbolization
It is interesting to note that cross-linguistic comparison only plays
a minor role in this listing. This fact corresponds to the methodical basis
of today’s Cognitive Linguistics. It is basically an interpretative framework
that describes direct or indirect relationship between cognitive events
and linguistic signs supported by the empirics of the sign in question
and the psychological events that can be documented when this sign in processed.
If you allow a somewhat polemic formulation, we may say that the old dictum
in especially French semantics, namely that chaque mot a son histoire,
is changed in Cognitive Grammar to chaque morphème a son évènement
cognitif. We are left with the impression that the strongly inductive orientation
of linguistic typology prevented this discipline from developing a theoretical
superstructure that would be capable of systematically incorporating possible
explanatory paradigms. Rather, we can observe two
different types of reaction towards the general challenge of cognitivism:
One the one hand, typologists are inclined to return to some rigid kind
of descriptivism, as formulated for instance by Bob Dixon in his Basic
Linguistic Theory. This rollback is accompanied by the trend to establish
some kind of categorial typology based on mere linguistic categories, a
trend that has its sources in Humboldt’s claim for a categorical encyclopedia
as well as in Bloomfieldian thinking and the post-war activities of the
Scoot de Linguistique de Paris, cf. (9). One the other hand, the explanatory
domain established by cognitivism is often accessed quite arbitrarily in
linguistic typology.
The linguistic discourse with the framework of typology
is hence characterized by a certain reluctance to systematic superstructures
that would formulate a deductive theory to approach the empirics of typology.
Still, the need of such superstructures – whatever they might be – seems
necessary to prevent typology from self-contained or even self-satisfied
descriptivism and explanatory haphazardness. The obvious progress that
has been achieved in the functional explanation of linguistic structures
calls for the systematization of this experience in terms of a unified
account. The best way to approach this task seems to be by creating some
kind of typology how and to which extent the explanatory domains
are accessed. Such a metatypology would be clectic and cumulative in
nature because we have to respect most of the empirical findings in linguistic
typology as well as the ways of how they are explained. A – if you allow
– philosophical superstructure of typology cannot and should not re-invent
the wheel but should try to propose a theoretical framework that makes
the many ends of typological explanation meet in an adequate format and
that is based on the most central axiom of linguistic explanation, namely
on derivationalism.
2. The Grammar of Scenes and Scenarios
Naturally, it would be obvious to use the term Cognitive Typology in
order to label such a framework. In fact, this term – already sometimes
used in informal talks – will serve as the title for an intended conference
the next year in Antwerp. However, these terms suggests that there is a
consensus in typology regarding the definition of what is finally covered
by cognition. Moreover, the term implies that there are different types
of typology, among them the cognitive approach. Yet, the specification
of the term typology normally yields at a given method applied to typology,
not at the theoretical foundation of the discipline itself, cf. (10), which
lists some of these terms. Another term that might be useful is holistic
typology. It refers to what is known as holistic cognitivism and explicitly
names the anchor chosen for typological explanation. Still, the term should
be used with great care, because it refers to an actual paradigm in cognitivism
that is rather susceptible to re-orientation.
Furthermore, we have to note that typology itself
rather represents a scientific method than the name for a program to explain
human language. Consequently, a possible superstructure of typology should
not be typology itself, but something that covers the typological approach
as well as other domains of linguistic derivationalism, among them those
mentioned in (8) of your handout.
In the second part of my paper I want to introduce
a theoretical framework that claims to contribute to the development of
a general superstructure for linguistic typology. I cannot work out all
the details of the proposed paradigm here, but will concentrate first on
some of its general properties only, before I will turn to some considerations.
The framework has been developed at the University of Munich and is labeled
a Grammar of Scenes and Scenarios. Its basic assumptions are worked out
in Schulze 1998; they will be further elaborated and evaluated in the forthcoming
volumes of the series Person, Klasse, Kongruenz with the help of a specific
empiric domain, namely that of the autochthonous East Caucasian languages
(cf. (11)).
The Grammar of Scenes and Scenarios, in short ‘GSS’,
should be understood both as a model for linguistic description and explanation.
Its deductive foundations are strongly related to what may be called Linguistic
Constructivism (cf. (12)). By this is meant that „language emerged from
brain structures which had been engineered for other functions, by co-opting
them from their previous domains. In essence, language ‘colonized’ the
earlier hominid brain; the end result is the modern human brain which is
built for language, but not specifically genetically programmed for it”
(Foley 1997:77). Hence, language systems are regarded as the routinization
of the emergent activities of the cognitive-communicative interface. They
are construed as a cognitive reality by human beings during the times of
language acquisition in accordance with the linguistic knowledge of their
instructors. It is important to note that GSS treats communication as a
cognitive procedure: It is represented by a phylogenetically older cognitive
substrate that is organized in a quasi-autonomous network or knowledge
system, cf. (14). Consequently, GSS tries to explain the grammar of a language
on the basis of the acquired cognitive and (cognition based) communicative
practice of an individual integrated into a collective. This practice is
dominated by massive hypotheses about the self-attachment to a collective.
It represents a strongly ritualized but construing interaction of the individual
with environmental or world stimuli which corresponds to the habitus of
a collective and which takes place in form of the tacit (poiematic) and/or
articulate (pragmatic) activation of an acquired (and traditional) knowledge
system. (Cf. 15)
Linguistic practice is thought to represent the
individual reaction to a collective communicative and cognitive standard
which itself is predominantly historical in nature. By this is meant that
the linguistic knowledge system of an individual and its instantiation
in a communícative community always reflects strategies of linguistic
adaptation that have been functionalized long before the individual has
acquired a given system. Hence, GSS argues that language as a ‘metaphysical’
phenomenon owns strong anachronistic features: It hardly ever meets the
immediate synchronic needs of information processing and communication.
It follows that functional and semantic aspects of language architectures
are mainly to be explained in a diachronic perspective, though the potential
to adopt newly established communicative and cognitive routines plays an
important role in the dynamic potential of language systems. The assumption
of an anachronistic ontology of language systems has an important consequence
for linguistic explanation: Contrary to some other cognitive approaches,
GSS does not establish a direct synchronic relationship between language
systems and cognition, cf. (16). Language systems and cognitive activities
are thought to be structurally coupled on the basis of a mainly diachronic
relationship that presupposes an adequate linguistic treatment of the data
in question.
To sum up this point: The theoretical framework
underlying GSS can be described as a basically diachronic model that owes
much to holistic cognitivism, constructivism, and pragmatism. The structural
coupling of adequate network components as well as the linguistic practice
emerging from this coupling shape linguistic paradigms: These activities
result in ‘language’ as a complex cognitive event – as an emergent activity
of this polycentric complex. Though GSS does not accept the modularization
of linguistic knowledge as a specific property of cognition itself, modularity
plays an important role in the GSS framework, too. This is because the
tendency of human beings to react on complex experience by singling out
specific components can lead to the construction of adequate symbols. Linguistic
modularity is thought to be such a construed symbol that is interpreted
as a cognitive reality and that is exploited by human beings to create
a cognitive hypothesis about the autonomy of language. People are used
to think of language as if it were autonomous just because it is the most
effective and most exploitable way to go through the experience of language.
Though GSS claims that this theoretical superstructure
is valid for all aspects of linguistic reality, it focuses on the architecture
of grammatical systems in order to meet one of the most central issues
of today’s typology, namely the typology of grammar. GSS hypothesizes that
the ontology of language is based on routinized mental constructions of
event experience (event images) that are structurally coupled with (tacit/articulate)
linguistic knowledge and activated in an intra/inter-individual communicative
(linguistic) context (cf. (17)). From this it follows that language ensures
the structural coupling of cognition and environment (as opposed to other
unilateral connections), which aims at a mediated way to share human experience.
Consequently, the most fundamental aspects of linguistic architecture should
be described in terms
of this functional relationship, cf. (18). However, it should be noted
that the linguistic paradigmatization of event images is not the only parameter
that organizes both linguistic knowledge and the architecture of grammatical
systems. On the one hand we have to respect the above mentioned anachronistic
aspect of language systems with respect to their cognitive condition, cf.
(19). On the other hand, linguistic systems are liable to be affected by
specifications of how communicative and linguistic practice is incorporated
into the dynamics of the habitus of a speech community, cf. (20).
Still, GSS claims that the cognition communication
interface represents the most basic domain of linguistic organization,
to which we have to add the Interface established by the CoCo domain and
the world knowledge base, cf. (21). The activities related to these two
interfaces strongly influence the way that people react on world stimuli
in terms of linguistically oriented event images. It is assumed that the
way of how such event images are processed by cognition is conditioned
by the communicative knowledge base: Linguistically processed pattern recognition
refers to a specific grammar that interprets and manipulates event images
in a communicative perspective (cf. (22)). In GSS, such event
images are called scenes.
Contrary to the tradition for instance of Joseph
Grimes and Ronald Langacker, scenes in GSS are regarded as a kind of cognitive
blueprint that is activated in pattern recognition. Hence, scenes do not
share any real world properties, but reflect the way in which real world
experience is construed on the basis of strongly idealized cognitive models
or cognitive hypotheses. The blueprints of scenes are thought to be part
of the evolution of cognitive and communicative behavior. Their basic structure
is constituted by the architecture of those cognitive domains that have
been involved in the emergence of scenic blueprint at all. It is assumed
that there is a functional iconicity between the neurophysiological architecture
of cognition and the architecture of scenes: Scenes cannot be processed
but within the general frame of cognition. Their blueprints represent engrammatic
structures that are stored in long term memory. The input of a world stimulus
activates procedures of picturing or re-presenting such engrammatic structures
in terms of what sometimes has been called scripts (Schank/Abelson), cf.
(23).
The CoCo interface establishes certain properties
of scenic blueprints that can be tentatively labeled as scenic universals.
By this is meant that the universal aspects of the CoCo interface forces
a universal design of scenes irrespective of the way how a given scene
is finally communicated. These universals are embedded into the general
schemes or cognitive hypotheses of human beings regarding the communicative
interpretability of world experience. Among them, we can describe (cf.
24) the degree of time stability, object and relational permanence, movement,
variation and change, localization, embodiment, and centrality. These universals
are additionally characterized by the fact that linguistic scenes are only
construed in a communicative context. Hence, we have to assume that basic
parameters of intra- or inter-individual communication and interaction
such as role swapping or role exchange also belong to the universal design
of such scenes.
All these universals constitute the kernel of prototypically
organized scenes. Such structures gain complexity because of two factors
(cf. 25): First, the universals can experience a different degree of particularization.
By this it is meant that higher level domains (cf. 21) as well as strategies
of grammaticalizing a scene establish a peripheric structure within the
prototypical organization of the scenic blueprints. Due to these particularized
blueprints for instance some universals become more active or functional
than others do. Second, scenic universals are liable to be a high degree
of metaphorical extension. The earlier such a metaphorical process has
become routinized and conventionalized, the more a resulting metaphor gains
the status of a quasi-universal. (26) Lists some of the most typical metaphorical
paths: Perhaps the most prominent one is that of causality that stems from
the universals of movement and change. Movement itself seems to result
from the embodiment of figure-ground-relations which themselves are part
of the overall scheme of internality and externally characteristic for
the process of embodiment. Movement and location are metaphorized as time,
which can interfere with causality and establishes modality.
Another universal property of the architecture of
linguistic scenes is that the original image schema or gestalt has to be
transferred into a serial sequence of information chunks that correspond
to the linearity of linguistic expressions. In GSS, this type of linearization
is called attention flow; a termed borrowed from the fragment of cognitive
grammar as proposed by Scott DeLeancey. However, attention flow has a specific
semantics in GSS: It refers to the fact that the cognitive constructive
reaction on a world stimulus already presupposes the anticipation of the
final linguistic output: Accordingly, the stimulus input is scanned with
respect to possible segments. This anticipatory behavior is called attention
in GSS. Though the construction of linearity itself is a universal scenic
feature, its instantiation is highly particularized.
Depending on the communicative context, but also on the higher level
domains mentioned above, scenes are grammaticalized to a very different
extent. The term grammaticalization has a doubled reading in GSS: On the
one hand, it refers to the traditional use of this term in the context
of grammaticalization research, that is in diachronic typology. On the
other hand it denotes the process of verbalizing a construed scenes. This
process is characterized by a complex interaction of different segments
of the cognitive network, cf. (27) which tries to outline some of those
factors that are involved in the procedure of grammaticalization. Unfortunately,
I cannot discuss all the signal components of this polycentric network
here. It should be sufficient to note that the relationship between the
blueprint domain of the scenic architecture and the
grammatical network that establishes a part of the linguistic knowledge
system is not regarded as a direct relation. The grammatical network does
not simply translate the scenic architecture into a linguistic format but
activates a complex relational network that plays the role of an interface
between the two components. It shapes the paradigmatic structure of the
grammatical network especially in a diachronic perspective, and restricts
the expression of scenic information to the formal extension of the grammatical
network.
This relational network is termed Operating System
in GSS. It refers to those parts of the grammatical network that constitute
the verbalization of scenes in a given speech community. Operating systems
represent basically tacit or poiematic linguistic knowledge systems and
guarantee an individual that it can express its communicative requirements
in a routinized and conventionalized form. They help to grammaticalize
the scenic architecture by anchoring a given scene in the grammatical,
lexical, and conceptual network components. Though Operating Systems tend
to be highly particularized reflecting the diachrony of linguistic and
communicative practices in a speech community, they have to observe some
general requirements that stem from the universal architecture of scenes,
cf. (28). The first column lists those CoCo parameters that condition and
dominate the scenic architecture. The second column gives the correspondent
architectural properties of the scene, whereas the third column refers
to some of those linguistic strategies that become relevant in the verbalization
of a scene.
It is important to note that all three domains are
prototypically organized. With respect to the scenic architecture we can
describe the following prerogatives that are located in the kernel of this
prototypical structure:
1. Scenes are always construed according to a figure-ground scheme.
Its metaphorical extension can be described as a vector relation between
cause and effect.
2. There are no empty scenes: Scenes share a minimal extension, which
is conditioned by the figure-ground relation. Hence a relation of one or
more participants (actors and actants) characterizes every scene. Relations
and actors can be secondarily masked or disguised, actors can be treated
as anonymous.
3. The population of a scene is theoretically unlimited in number,
however, the CoCo interface as well as principles of perception and information
processing restrict the number of possible actors (usually, not more than
three to four actors per scene).
4. Scenes are always located in (system) space and (system) time.
5. Every scene shows a perspective arrangement: This arrangement can
by static or dynamic.
6. Every scene shows a notion of centrality: The most central perspective
is taken by the construing individual which can delegate this role to either
other Speech Act Participants (role swapping) or to actors in a given scene.
7. Every scene is structured according to the attention flow.
The organization of operating systems copies this prototypical arrangement to a certain extent. We can observe that the Accusative Ergative Continuum plays the most central part in the verbalization of scenes. The AEC dominates for instance figure-ground relations as well as the cause->effect vector, attention flow, time and space allocation, perspectivization, and – partly – centralization. Consequently, GSS treats the AEC as the most prominent aspect of a given operating system. The AEC itself can be characterized as a linguistic reaction upon the dynamics of the cause->effect vector in correlation with other parameters of the scenic architecture. It is claimed that every linguistic paradigm that becomes effective with respect to the verbalization of the cause->effect vector is subjected to the AEC. The AEC allows the description of the paradigmatic, syntactic, pragmatic, and semantic behavior of a given linguistic structure in view of the weighting of actors in a scene, cf. (30). The more weight is layed upon the C domain, the more the paradigm behaves accusative-like. And vice versa: The more weight is layed upon the E domain the more is operates upon an Ergative-like schema. Naturally, we cannot claim that an operating system behaves totally accusative or Ergative. Rather we should assume that quite often the single linguistic units are differently located on the AEC – depending on the functional value a given paradigm has with respect to the scenic architecture. (31) Lists some of those linguistic ‘categories’ that are sensitive for the AEC. Criteria of cognitive and communicative dominance as well as aspects of co-paradigmatization or structural coupling among the single paradigms help to describe the general dynamics within the polycentric structure of the operating system in question.
If you again turn to (12) in your handout, you can see that my paper
has been organized along the four basic explanatory domains of GSS. Namely
the general theoretical framework, the Grammar of Scenes and Scenarios,
its grammaticalization in terms of the operating system, and – finally
– the organization of the operating system itself. Among the many questions
that remain probably the most important is whether such a concept helps
to develop a superstructure for language typology that is powerful enough
to take into account the results of formal and functional typology. As
well as current approaches to language theory in the framework of derivationalism.
The fact that I did hardly ever use the term typology within this presentation
of GSS may question the assertion that this framework is interested in
typological data at all. But we have to bear in mind that typology itself
on the first hand is nothing but the name for a methodical approach that
claims to establish linguistic types for further evaluation. These types
correspond to the observed diversity of linguistic systems. Hence, typology
has acquired a connotation that is strongly connected with linguistic variation
rather than with universals of human language. GSS accepts this methodical
basis and heavily relies upon its descriptive and analytic output. GSS
is a typological theory, because it is particularly interested in the process
of linguistic diversification as a source to determine those parameters
that are responsible for it. GSS is a typological theory also, because
it tries to establish a typology of just those domains that are thought
to constitute language as well as languages. Just as the diversity of linguistic
paradigms calls for an adequate typological description, the systems that
control the linguistic output have to be described in terms of their possible
particularization. But contrary to many current approaches in language
typology, GSS tries to integrate this kind of typological thinking into
the broader context of universals of human communication and cognition.
It aims at explaining the process of particularization on behalf of universal
parameters, because you can only account for this process if you accept
that everything that establishes a type via particularization must result
from something more universal. Consequently, if you want to pursue the
third level of typological research called typology proper (see again (3)),
namely to the explanation of linguistic types, you have to turn to a model
that makes proposals concerning the conditional nature of these types.
The way to simply call all those features universals that are common to
all languages has turned out to be a dead lock. In fact, the number of
such features is considerably low and often enough we find nothing stated
but pleonasms. GSS argues that universality can hardly be described in
terms of paradigmatic features. Rather, it is the behavior of linguistic
paradigms with respect to their derivational basis that bears some notions
of universality. In this, GSS differs considerably from traditional typological
approaches, cf. (32), which documents the basic structure of argumentation
in GSS. The assumed explanatory power of GSS perhaps lies in the fact that
it encompasses both inductive and deductive methods without giving technical
priority to either one. It relies on the assumption that every linguistic
unit – be it located on a micro- or on a macro level – in itself carries
the whole world of cognitive events that are activated when this unit is
produced. However, GSS cannot be regarded as a variant of Cognitive Linguistics
because such a tradition implies that it interprets linguistic data from
a cognitive perspective. From this it would follow that other perspectives
are a valid option, too. GSS is more radical: It claims that language is
nothing but part of the cognitive domain. Language does not exist beyond
this domain, just as all human activities are finally processed by cognition.
This basic claim of constructivism unmasks the term Cognitive Linguistics
as yet another pleonasm.
Let me finally turn again to the title of this paper:
If Cognitive Linguistics is a pleonasm than we have to change the title
to either Linguistics meet typology which sounds more than absurd, or we
have to reformulate it as Cognitive sciences meet language typology. The
term meeting, however, suggests that it is a first step towards integration.
Hence, we arrive at the name for a scientific program which can described
as language typology becomes integrated into Cognitive sciences. It presupposes
that the Cognitive sciences open their windows to the results already achieved
by the high and very sophisticated standard language typology. Naturally,
this claim has to work in the opposite direction, too. The framework of
GSS tries to propose such a path for language typology as well as for language
theory in general. Whether it can contribute to the progress of linguistic
research or whether its assumptions are nothing but „straws in a wind that
will die down or prove to be blowing elsewhere” (to quote Matthews 1993:48),
only the future will tell.
Munich, 18.6.1999 / 14.9.00 (copyright W. Schulze)