I
Basic
Issues
1.
SYSTEMS AND MODELS
1.1
In many countries, the increasingly prominent role of text linguistics in the
discipline of language study appears to signal a “paradigm shift” in the
sense of Thomas Kuhn (1970). The older preoccupation with demonstration
sentences isolated from communicative contexts is yielding to a new concern for
the naturally occurring manifestation of language: the TEXT. Language
occurrences may have the surface format of single words or sentences, but they
occur as texts: meaningful configurations of language intended to
communicate. The implications of this shift of investigation are far-reaching
indeed. We are not simply moving from the exploration of shorter toward longer
language samples. We are also replacing an emphasis on abstract forms with an
interest in the UTILIZATION PROCESSES of language (cf. Hörmann 1976; H. Clark
& E. Clark 1977).
1.2
The traditional restrictive confines of linguistics are gradually giving way to
concerted interaction with other language-related disciplines: psychology,
sociology, philosophy, computer science, semiotics, cybernetics, education, and
literary studies. Unless linguistics is to disappear as a separate field
(envisioned by Yngve 1969), it should become the pivotal science of discourse
and communication envisioned by so many astute researchers (e.g., Lévi-Strauss
1958: 37; Dundes 1962: 96; Hymes 1962: 9; Piaget 1966: 25; Hartmann 1970: 53;
Maclay 1971: 180f.; Jakobson 1973; Koch 1973/74: xi). The broad usefulness and
applicability of linguistic theories and methods would then figure as a
prominent goal and not, as in past times, an incidental by-product, or even a
misunderstanding.
1.3
In this wide framework, text linguistics would constitute the verbal domain of
semiotics, dealing with the entire range from one-word texts (e.g., ‘Fire!’)1
[Throughout the book, I enclose linguistic samples cited within the
running text in single quotes; punctuation is included inside only if part of
the sample to texts as vast as The Divine Comedy. The decisive trait of
the text is its OCCURRENCE IN COMMUNICATION (Hartmann 1964), where it is
produced by a single participant within some temporal limits (cf. Weinrich 1976:
187). A set of mutually relevant texts can be said to constitute DISCOURSE,2
[2. The term “discourse analysis’ has been variously used for
beyond-the-sentence linguistics in general (Harris 1952) and for the study of
conversation in particular (Coulthard 1977). In my scheme, both of those domains
are only parts of a science of texts as actual communicative occurrences.]
a progression of occurrences that may be continued at a later time (cf.
van Dijk 1977a). The total constellation of mutually relevant discourses in a
group or society can be called a UNIVERSE OF DISCOURSE (cf. Coseriu 1955-56;
Pike 1967: 596; van Dijk 1977a: 127).
1.4
Human language is so complex in its organization and so diverse in its
manifestations that the science of linguistics naturally remains in continual
evolution. The linguist faces a formidable overabundance of data ranging from
observable, face-to-face speaking to abstruse mathematical and philosophical
speculations on language. In its early stages, linguistics was compelled to be
highly selective and reductive in its treatment (Uhlenbeck 1973: 107; Grimes
1975: 3). Continued progress has rendered much of this attitude superfluous,
though debates over admissible issues still rage. As our scope expands, we
approach the time when linguistics may be able to meet the demands that society
imposes upon a science of language (cf. Hartmann 1970).
1.5
However much data a researcher may gather and evaluate, data can be significant
only with respect to the COGNITIVE INTERESTS of a discipline (cf. Kuhn 1970;
Schmidt 1975): commitments to seeking certain kinds of knowledge. Especially in
linguistics, what constitutes worthwhile data, or how data should be treated, is
by no means self-evident. The cognitive interests expressed in this volume are
devoted to texts as vehicles of HUMAN ACTIVITIES — a notion already envisioned
by many founders of linguistics (e.g. Malinowski 1923; Jespersen 1924; Bühler
1934).
1.6
Using a term made current by Carl Hempel (quoted in Stegmuller 1969: 205), we
can define the scientific treatment of data as SYSTEMIZATION: the imposition of
a system upon obtainable evidence. A SYSTEM is considered to be a unity of
mutually relevant ELEMENTS3 [3.I
use the term ‘element” for any item whose occurrence or use is governed by
SYSTEMIC (pertaining to a system) principles.]
whose FUNCTIONS are determined by their respective contributions to the
workings of the whole. To account for data, researchers build a SYSTEMIC MODEL
whose operations might yield such data (on the notion of the model, see Hartmann
1965; Gülich & Raible 1977). The correlation of the model with an empirical
domain is regulated by BRIDGE PRINCIPLES (Hempel 1966) that state the degree of
APPROXIMATION between the model and the domain (cf. Apostel 1961). Ideally,
scientific progress continually reduces the degree of approximation and makes
the model a more exact representation. However, the object of study may itself
be FUZZY: characterized by structures and operations that are probabilistic and
not exhaustively delimited. Such is the case with natural language
communication.
1.7
The view of language as a system is well established (see for instance Saussure
1916; Weinreich 1954; Firth 1957; Halliday 1967a, 1969; Heger 1971; Laboy 1971;
Winograd 1972; Berry 1977; Clark & Clark 1977; van Dijk 1977a). The systems
approach implies cognitive interest in the DYNAMICS of an entity (Hartmann
1963a, 1963b), such as CONTROL and INTERACTION among elements. Systems theory,
which has become a discipline in its own right (cf. Boulding 1956; Bertalanffy
1962; Buckley [ed.] 1968), has found acceptance in research areas as diverse as
psychology, sociology, behavioral science, design engineering, information
science, computer science, factor analysis, thermodynamics, mathematical
topology, and many more. The prevalence of the systems approach is heuristically
enriching in allowing models to be shared and borrowed among disciplines. To be
sure, borrowed models have crucial effects upon research methods, and must be
applied with caution.
1.8
Language is initially given as a MANIFESTATION: an occurrence or set of
occurrences at least partly accessible to apperception (cf. Stegmüller 1969:
93). The observable aspects interact with non-observable ones in intricate and
diverse ways. The total picture of a language must be gradually assembled via a
sequence of systemizing tasks, for example:
1.8.1
IDENTIFYING a manifestation, its constituents, or its environment;
1.8.2
GENERALIZING about related or relevant manifestations;
1.8.3
DESCRIBING a set of manifestations methodically;
1.8.4
EXPLAINING the existence or occurrence of manifestations;
1.8.5
PREDICTING manifestations under statable conditions;
1.8.6
RECONSTRUCTING artificial correlates of the manifestations;
1.8.7
MANAGING the occurrence of manifestations.
1.9
This list of tasks is arranged in what 1 consider the order of increasing
difficulty. Accomplishment of a given task presupposes prior achievement of
those above it. In practice, however, we must often work provisionally without
these prerequisites. For example, public education may require the management of
language in absence of any thorough explanatory or predictive account.
1.10
In order to establish itself as a discipline, linguistics initially dissociated
itself from the prescriptive ambitions of traditional grammar and sought to
develop some objective, reliable tools for the more basic tasks of
identification, generalization, and description. This early phase, usually
entitled “descriptive linguistics,” attained a sufficiently rigorous
methodology to uncover the grammars of numerous hitherto unrecorded languages,
even with no prior knowledge of their structure (especially via the
“tagmemics” developed by Kenneth Pike [1967] and Robert E. Longacre [
1964]). Emphasis was understandably placed upon the language aspects most
accessible to observation: sounds, forms, and arrangements of utterances. The
treatment of non-observable aspects, such as communicative strategies or
psychological processes,4
was informal and intuitive. Communication as a human activity was not
viewed as a major object of study in its own right.
1.11
The “generative” approach to language study embarked on the more arduous
tasks of explanation, prediction, and reconstruction (beginning with Hjelmslev
1943; compare Chomsky 1957). Generativists borrowed heavily on formal logic to
build an idealized model of human language with stringent restrictions on the
object of study. The divergences between their model and the real data were
sufficiently strong in some domains to cause an apparent antagonism between
“model-oriented” and “data-oriented” studies (Liefrink 1973). However,
the expansion of linguistics to new tasks remains an enduring contribution of
the generative approach (cf. Dingwall 1971).
1.12
If models mediate between what we can apperceive and what we want to explain
(motto attributed to Anaxagoras by Gulich & Raible 1977: 14; cf. Wagner
1974: 150), then the following situations would justify reliance on models
(adapted from Apostel 1961):
1.12.1
when no known theory exists for a domain;
1.12.2
when a theory is known, but too complex to allow solving problems with currently
available techniques;
1.12.3
when a theory is known and partly confirmed, but still incomplete;
1.12.4
when new research permits the correlation or integration of two or more known
theories;
1.12.5
when the objects of investigation are too large, too small, too remote, or too
arduous to allow direct observation and experimentation.
1.13
All of these situations obtain to some degree in linguistics:
1.13.1
Some areas are still without a workable theory, such as the interface of
language with emotional states of communicative participants. Peter Hartmann
(personal communication) comments that the notion of ‘observation’ in these
early discussions was somewhat out of place. The actualisation of communicative
strategies guarantees the potential for objective study, whether or not we are
looking directly at the data in its most detailed form.
1.13.2
Some theories are available, such as Montague’s (1974) grammar, but are too
intricate to be useful in solving empirical language problems, at least with
current technology.
1.13.3
The “generative’ approach to language remains incomplete until it can
explain how texts are actually produced and understood by humans.
1.13.4
Research on texts demands the integration of theories from many areas: sentence
grammar, philosophy, computation, cognition, planning, and action. 1.13.5 Some
aspects of language are too large (e.g. the totality of discourse in a whole
society), too small (micro-impulses of nerve cells during language processing),
too remote (storage of knowledge in the mind) or too arduous (relating every
minimal feature of utterances to its social, psychological, and historical
evolution) to be pursued in direct experimentation.
1.
14 To sort out the regularities of language from the accidental details, one can
distinguish between the SYSTEMIC aspects of language, styled “langue” or
“competence,” and the seemingly accidental or irrelevant aspects, designated
“parole’ or “performance.” During the evolution of the discipline, the
borderline between the two aspects shifts as domains once thought to be
fortuitous are discovered to have a systemic nature after all. For instance, the
sentence was assigned to “parole” by Saussure (1916: 172), only to become
the primary entity of “competence” in transformational grammar since Chomsky
(1957). I shall argue in 1.5 that “textual competence” encompasses a
markedly different domain from sentence competence (cf. also IV. 1.24).
1.15
The respective limits of “langue” or “competence” influence and are
influenced by the models and methods in use. Descriptivists broke down their
sample evidence by extracting LEVELS of MINIMAL UNITS to be classified into a
TAXONOMY: a scheme for sorting elements by distinctive features. If each level
of minimal units was thought to be a system of mutual oppositions, the entire
repertory of each system had to be exhaustively assembled, for example sounds
(phonology) and forms (morphology); exhaustive treatment of meanings or
situations was deemed impossible, and those domains were set aside. Later, the
generativists preferred to begin at the other end with a GRAMMAR as a set of
rules stipulating what does or does not belong to the language. The problem of
exhaustiveness was suspended by postulating that all complex entities (however
many there might be) could be derived from a limited set of simple ones
(kernels) by using the proper rules. The rules were designed so as to produce an
infinite set of sentences.
1.16
The generative approach is vastly more ambitious than the descriptive one, since
it must not only systemize all occurrences: it must also preclude all
non-occurrences (McCawley 1976). Strictly speaking, it is not a grammar of
occurrences at all, because it professes to deal only with abstract potential.
The empirical verification of such a grammar can be a major difficulty. Outside
the relatively small range of obvious, uncontroversial cases, people have
trouble judging which utterances their native language should allow (cf. Labov
1966; Lakoff 1969; Carden 1970; Heringer 1970; Wedge & Ingemann 1970; Ringen
1975). Grammaticalness judgments are supposed to apply only to structure, not to
context. But structures never occur naturally without context, and language
users therefore do not possess the expertise needed to make consistent
judgments. Informants are, in reality, trying to imagine possible contexts for
each sample (Uhlenbeck 1973: 42; McCawley 1976: 155; van Dijk 1977c; Snow &
Meijer 1977). The designation of “grammatical” is awarded to banal sentences
whose occurrence is easy to imagine for everyone (Householder 1960: 340). For
less banal cases, opinions are unstable and inconsistent. Heringer (1970) found
that providing a context for such a simple sentence as ‘John left until 6
P.M.’ elicited a change in people’s grammaticalness judgments of no less
than 40%! People are still less able to decide about elaborate utterances, such
as those which Robinson (1975: 141ff.) contests against Chomsky (1972).
1.17
The lack of empirical verification procedures for large sections of a language
theory can lead to disquieting tactics (cf. Beaugrande 1979k):
1.17.1
Circularity of proof: the correct rules are those which generate only
grammatical sentences; grammatical sentences are those generated by the rules
(cf. critique in Dik 1967; Uhlenbeek 1973).
1.17.2
Intuitions: the linguist presents demonstration sentences which are (in
his or her own intuition) grammatical or ungrammatical (criticism in Dik 1967:
372; J. Anderson 1976 :69; Schiesinger 1977: 210). The linguist becomes the
informant and is free to exclude unfavorable examples at will (cf. Rieser 1978:
8), thus sacrificing both objectivity and generality.
1.17.3
Appeals to the competence/performance distinction: all data that the
theory cannot treat are shunted off into the domain of “performance” and
excluded from consideration as non-issues (criticism in van Dijk 1972a: 314). An
egregious and anachronistic illustration is Dresher and Hornstein’s (1976:
328) claim that “a study of competence abstracts away from the whole question
of performance, which deals with problems of how language is processed in real
time, why speakers say what they say, how language is used in various social
groups, how it is used in communication, etc.” A language user without all of
that knowledge could be called “competent” only by a bizarre perversion of
the whole notion in its commonsense usage.
1.18
If a language model can do no more than assign structural descriptions to
sentences, then it deserves the old designation “structural descriptive
linguistics” rather than “generative.” To earn the latter designation, the
model should suggest how utterances might actually evolve (J. Anderson 1976:
118; Simmons 1978: 2). Only then can we reasonably conduct empirical tests and
agree on rational standards for evaluating, verifying, and accepting one account
of language over any other.
2.
LEVELS IN MODELS OF LANGUAGE
2.1
A spoken or written text in English could appear to be or to consist of various
things. One observer might notice a stream of sounds or forms following each
other in the real time of speaking or of moving from left to right on a page.
Another might notice that the text is intended to embody knowledge and meaning.
Still another might notice that the text could be a vehicle for someone to get
something done or to reach a goal. Each of these observers would be apperceiving
a single, simultaneous aspect of the text: one of its LEVELS.5 [5.
The term ‘level” has been used indiscriminately in the past, often being
conflated with such notions as ‘rank.” I consider a “level” to be the
total aspect of a participating language system; a ‘rank” is a unit of a
given dimension in a hierarchy of size (e.g. word, sentence, etc.).] It seems
reasonable that language science should attempt to extract and systemize these
levels as a proper domain of investigation.
2.2
In its early phases, linguistics proceeded on the assumption that levels should
be systemized independently of each other (e.g., Trager 1950). That outlook
seemed to be successful for the description of sounds, though, as Kenneth Pike
(1967: 362f.) notes, it was not fully upheld even by its defenders. Later, acute
problems emerged when the borderline between morphology and syntax came into
view. A distinction was drawn between the paradigmatic aspect that determined
what items might fill a slot, and the syntagmatic aspect that determined the
sequence of slots themselves.
2.3
The independence of syntax from meaning was maintained with considerable vigor
(cf. Harris 1951; Chomsky 1957). Harris himself allowed for an expedient to use
meaning as a short-cut for analysing language, provided that a purely formal
analysis of the distribution of language items would arrive at the same result.
In effect, Harris postulated that an item’s meaning is the sum of all
positions it can occupy in usage. This postulate is not in itself unreasonable,
but for a model of either linguistic analysis or human language activities, it
is unworkable. Meaning would remain undiscoverable until we had catalogued all
distributions of a given item.
2.4
Rather than upholding the separation of syntax from meaning, Harris’ postulate
tends to suspend it. Meaning and syntax must interact in order for language
items to have a given distribution. I shall argue, however, that we should go
further and investigate the PROBABILITY of occurrences in systematic
environments. The “well-formedness” (i.e., conformity with the grammar) of
language sequences is not, taken by itself, a sufficient principle (cf. 11.
2.36ff.; IV.1.24).
2.5
Although reaffirming the independence of syntax from meaning, Chomsky’s (1957)
“transformational grammar” sought to escape the unworkability of Harris’
postulate of distribution. Instead of analysing the distributions of language
items as such, Chomsky undertook to set up an abstract rule system which could
produce all allowable distributions of a language. Attention was transferred
from the analysis of extensive samples toward the construction of rules. In
essence, this transfer did not simplify linguistic research. Every
counter-example to the previous rules called for new rules — a factor making
the transformational model immune to falsification as a theory.
2.6
In semiotics, it is traditional to subsume all aspects of formal arrangement
under the notion of SYNTAX, and all aspects of meaning under that of SEMANTICS;
the use of language was subsumed under PRAGMATICS. To deal with an entire
language, the “transformational” approach began with an autonomous set of
syntactic rules; semantics was treated as an after-the-fact ‘interpretation’
of syntactically produced strings. In some models, pragmatics was simply added
as a further phase of “interpretation” (e.g. van Dijk 1976). Such an
approach was obliged either to ignore the interaction of these three factors in
the actual production and comprehension of utterances, or to reconstruct them
all in terms of arbitrary syntactic rules. An alternative account in which
meaning was given the key role from the outset was introduced as ‘generative
semantics” (see. 11. 1.6). Quite aside from the detailed issues of rule
constructing, this controversy pointed up a basic question concerning the
building of language models. I shall examine the question here from a
systems-oriented viewpoint.
2.7
In systems theory, we can distinguish the approaches of MODULARITY and
INTERACTION (cf. Sussman 1973: 12f.; Winograd 1975: 192). The reliance upon
formal logic and mathematics in generative grammar fosters modularity, in which
system components are substantially independent, and operations are cumbersome
(cf. Levesque & Mylopoulos 1978: 2). My outlook here will be directed toward
interaction, without which the utilization of text would simply not be
operational (cf. 11. 1; Walker [ed.] 1978).
2.8
Imagine for a moment a different kind of language model. We might start out with
these two well-known levels:
2.8.1
SYNTAX PROPER is concerned with the abstract patterns and sequences which the
grammar of a language stipulates independently of context.
2.8.2
SEMANTICS PROPER is concerned with “the relations between signs or symbols and
what they denote or mean” (Woods 1975: 41). The repertory of signs and symbols
with a statement of their meanings is contained in the LEXICON. If lexical items
are defined according to their content, we have INTENSIONAL MEANING (e.g.,
‘blue’ is the color lying in the spectrum between green and violet); if
items are defined by their REFERENCE to entities, we have EXTENSIONAL MEANING (e.g.’blue’
is the property shared by all blue things in the world). The standards for
judging the correctness of statements about some world and for combining
statements in that perspective are set forth as TRUTH CONDITIONS. The extent to
which a reference encompasses an object or class of objects is the issue of
QUANTIFICATION (e.g. ‘every person’ or ‘all persons’). The probability
or necessity of a statement regarding some world is its MODALITY.
2.9
In the above definitions, syntax and semantics proper are indeed independent of
each other. Formal sequences can be envisioned before deciding what specific
lexical items might fit into them; and lexical meaning need not commit an
expression to appearing in a given slot of a sequence. Yet no utterance could
ever be produced without making these decisions and commitments, and none could
be understood without recovering them. It follows that syntax proper and
semantics proper as previously set forth are components of logical languages,
but not of natural languages in use. Instead, let us envision two different
levels of language use:
2.9.1
The SEMANTICS OF SYNTAX is concerned with how people utilize formal patterns and
sequences to apply, convey, and recover knowledge and meaning. For example,
noticing a noun-verb sequence might give rise to an expectation that an agent
and an action are being expressed (cf. III.4.16.1).
2.9.2
The SYNTAX OF SEMANTICS is concerned with how concepts like agent, action,
state, attribute, etc. are connected to yield the total meaning of a text.
Semantics of syntax has a more predominant linear or sequential organization
than does syntax of semantics. For instance, an action might be linked to an
agent, a time, a location, a cause, and so on, while various linear arrangements
would be possible for expression (cf. VII.2) (on grammars without fixed
linearity, cf. Petöfi 1972).
2.10
These interfaced levels are not wholly novel here (compare Ihwe 1972: 339;
Schank 1975b: 14f; Rieser 1976: 13). Their function is to co-ordinate operations
people perform when utilizing meaning in connected utterances. I shall develop
this direction by pursuing the notions of SEQUENTIAL CONNECTIVITY: how elements
are arranged in the surface text; and CONCEPTUAL CONNECTIVITY: how underlying
concepts and relations are put together.6
[6. Compare the notion of ‘sense constancy’ in Hörmann 1976, ch. 7.]
The interaction of these two is controlled by MAPPING PROCEDURES (cf. Goldman,
Balzer, & Wile 1977). The characteristic mapping procedures selected to
produce a text yield the STYLE of the text.
2.11
When each language level is systemized, the entire language appears as an
INTERSYSTEM the workings of which depend on the interaction of participating
systems (cf. Halliday 1969; Berry 1977; Dressier 1979). Each system has INTERNAL
CONTROLS that regulate the availability of options and the allowability of
combinations; and EXTERNAL CONTROLS which regulate that system’s interaction
with other SYSTEMS.7 [7.
In his general systems theory, Luhmann (1970) stipulates that every system must
have a “differentiation of internal and external,” i.e. be distinguishable
from its environment. Linguistics has been concerned with differentiating sets
of system elements at the expense of operations, functions, and controls. As J.
Anderson (1976: 80) notes, control has hardly been studied in cognitive
theories.] Both kinds are indispensable to the production and utilization of
texts, but external controls have received little attention in conventional
linguistics.
2.12
Issues of control were largely assigned to the poorly explored domain of
pragmatics, dealing with language use. In that line of reasoning, pragmatics
becomes “meta-syntax” and “meta-semanties,” rather like a self-awareness
of decisions about arrangement and meaning. However, in order to attain a
workable design, each system should have at least its own essential controls
built right into it. Pragmatics is properly the domain of the human activities
of PLANNING texts as vehicles of PURPOSIVE ACTIONS directed toward GOALS
(Beaugrande 1979b). The theory of texts accordingly requires a different triad
of domains than the old semiotic scheme:
SYNTAX
=> SEQUENTIAL CONNECTIVITY
SEMANTICS
=> CONCEPTUAL CONNECTIVITY
PRAGMATICS => ACTIONS, PLANS, GOALS
Each
domain is subject to relevant controls during communication. The discrete items
occur within a CONTINUITY which arises from the DIRECTIONALITY of CONTROL FLOW.
Accordingly, we need a dynamic outlook for investigating not only the presence
of structures in texts, but also the operations that can create, build, and
utilize structures (cf. Hartmann 1963a; Mukarovsky 1967: 11; Woods 1970;
Winograd 1972; Koch 1976). If we define STRUCTURE as a relation between at least
two systemic elements in occurrence, it is clear that a theory of language use
should be centered upon the notion of CONNECTIVITY.
3.
TEXT VERSUS SENTENCE
3.1
Nearly all accounts of language structure set forth since classical antiquity
have relied decisively on the notion of the SENTENCE. It is disquieting that
this basic entity has been vaguely and inconsistently defined, even up to the
present (O’Connell 1977; Glinz 1979; Beaugrande 1999). Different criteria for
‘sentencehood’ have persisted without being explicitly recognized as
divisive rather than unifying
standards, for example: (1) the expression of a “complete thought” (see Ivic
1965: 20); (2) a sequence of speech units followed by a pause (see Gardiner
1932: 207; Goldman-Eisler 1972); (3) a structural pattern with specified formal
constituents (cf. Harris 1951; C. Fries 1952; Chomsky 1957). The functional
implications of each of these criteria are radically distinct from the others.
Empirical research makes it plain that people disagree with each other’s
judgments about what constitutes a sentence. When speech pauses were consulted,
“many segments identified by this study as sentences would not be considered
separate sentences by other criteria” (Broen 1971: 30). A still graver problem
is that the boundaries of utterances are often marked by non-linguistic signals
(Hörmann 1976: 329).
3.2
As O’Connell (1977) notes, linguists usually accept the sentence as a basic
entity a priori, thus bypassing the methodological difficulties. In
transformational grammar, language is in principle defined as a set of
sentences. Whatever is not found as a sentence (e.g. a mere noun [Lees 1960]),
must be converted into one by transformations and derivations. The sentence was
inconsistently treated not only as a grammatical pattern, but also, whenever
occasion arose, as a logical statement. But this duality is a property of
logical languages, not natural languages. Entities such as “argument” and
“predicate” are definable in terms of logic, but “noun phrase” and
“verb phrase” are purely grammatical entities.
3.3
It appears that linguists tended to confuse optional mapping with obligatory
mapping. The sentence provides no more than a grammatical format into which
semantic and pragmatic unities can be mapped; some linguists treated the
sentence as the format into which these unities must be mapped. This
practice undermined the proclaimed autonomy of syntax, because many attributes
assigned to the sentence actually belong to semantic and pragmatic unities.
Consequently, the question of how humans decide
what to map onto what could hardly be raised. We can readily observe that people
can make a wide range of decisions about syntactic formatting (cf. VII.2). As
long as linguistics presupposes the sentence at the outset, such facts are
difficult to treat; the linguist is compelled to retreat into a remote,
reductive version of “ideal competence.” An impasse has arisen beyond which
linguistic theory cannot advance, because the most basic concepts are short-
circuited across each other, rendering many vital realities of communication
inadmissible issues.
3.4
I assert that the multi-level entity of language must be the TEXT, composed of
STRETCHES OF TEXT which may or may not be formatted as sentences. I would cite
the following essential distinctions between text and sentence:
3.4.1
The text is an ACTUAL SYSTEM, while sentences are elements of a VIRTUAL SYSTEM,
as I shall explain in 1.4.8 [8. Hence, it is pointless to
debate whether sentences can have meaning inside or outside contexts (Bever vs.
Olson, cited in Kintsch 1974: 15). The meaning belongs in any case to the sentence-formatted
text; the format is at most a means to provide certain signals about the
configurations of meaning (cf. III.4.16ff.).]
3.4.2
The sentence is a purely grammatical entity to be defined only on the level of
SYNTAX. The text must be defined according to the complete standards of
TEXTUALITY as elaborated in section 1.4.8
3.4.3
In a text, the grammatical constraints imposed upon abstract sentence structures
can be OVERIDDEN by context-dependent motivations.9 [9. The phenomenon of ELLIPSIS (cf. V.6) is a good
demonstration.] For example, elements easily recoverable from a situation via
sensory apperception can be omitted or truncated by the speaker without damaging
the “cornmunicativity”of the text. Grammaticality should be treated not as a
law, but as a DEFAULT: a standard assumed in absence of specific indications; or
as a PREFERENCE: a standard to be selected over others when various options are
open (cf. the notions of “default” and “most likely case” in Collins,
Brown, & Larkin 1977: 17; and “preference” in Wilks 1975b, 1978).
3.4.4
The distinction between “grammatical” and “non-grammatical” is a binary
opposition if one has an accurate and complete sentence grammar (R. Lakoff 1977)
(which is not the case so far): one decides if a given entity is a sentence by
matching it against the sequences produceable by grammatical rules. But the
distinction between a “text” and a “non-text” is not decided by any such
mechanical checking. Texts are ACCEPTABLE or NON-ACCEPTABLE according to a
complex gradation, not a binary opposition, and contextual motivations are
always relevant. It is well known, for example, that some respected literary
texts are and must be beyond the range of any reasonable grammar (cf. S. Levin
1962; Thorne 1969; van Dijk 1972a, 1972b) (cf. IX.7.1ff.). Since the text is
defined on the basis of its actual occurrence, the notion of a “non-text” is
a marginal concern.10 [10.
The use of counter-examples, many of them bizarre and contrived, has been
exaggerated in linguistic arguments, doubtless due to the striving for a
categorical (context-free) well-formedness grammar. Counter-examples do not
overthrow important regularities of a language (Wilks 1975a); cf. the samples
indexed in footnote 14.] Linguists who deliberately set out to construct
non-texts are no longer participating in communication, and thus are not likely
to explain the normal workings of the latter.
3.4.5
A text must be relevant to a SITUATION of OCCURRENCE, in which a constellation
of STRATEGIES, EXPECTATIONS, and KNOWLEDGE is active. This wide environment can
be called CONTEXT; the internal structuring of the text constitutes the
CO-TEXT’ (on this distinction, see Petöfi 1971b, 1975a; Petöfi & Rieser
1974).11 [11. Petöfi’s use of these terms is more
rigorously defined. “Co-text” is said to subsume: grammatical components,
syntax, intensional semantics, morphology, and phonology (or, in written texts,
graphemativs); “context” subsumes extensional semantics, and the production
and reception of texts (Petöfi & Rieser 1974: vi; Petöfi 1975a: 1).]
One can design sentences, on the other hand, that might never occur
spontaneously, being too long, too complex, too heavily embedded, too trite or
inane semantically, or too pointless pragmatically. The rules for abstract
sentence formation alone cannot stipulate some maximum length or complexity
beyond which a sequence ceases to be a sentence.12 [12. Hence,
sample (194) in VII.2.25 is grammatical, but hardly acceptable in
communication.]
3.4.6
A text cannot be fully treated as a configuration of morphemes or symbols. It is
the manifestation of a human ACTION in which a person INTENDS to create a text
and INSTRUCTS the text receivers to build relationships of various kinds. As
such, INSTRUCTION figures as the eliciting of processing actions (cf. Schmidt
1971c, 1971d, 1973; Weinrich 1976). Texts also serve to MONITOR, MANAGE, or
CHANGE a SITUATION (cL Kummer 1975; VI.4). The sentence is not an action, and
hence has a limited role in human situations; it is used to instruct people
about building syntactic relationships.
3.4.7
A text is a PROGRESSION between STATES (Chafe 1976. 27£; Fowler 1977: 77): the
knowledge state, emotional state, social state, etc. of text users are subject
to change by means of the text (cf. ‘epistemic change” in van Dijk 1977a:
194). The production and comprehension of a text are enacted as progressive
occurrences. At each point in those progressions, CURRENT CONTROLS apply which
need not be identical with abstract formation principles. For example, the
controls upon beginnings of texts differ from those upon continuations or
endings (cf. Harweg 1968b). In contrast, sentences are to be viewed as elements
of a STABLE SYNCHRONIC SYSTEM (i.e. a system seen in a single, ideal state free
of time), so that controls apply CATEGORICALLY (obligatorily and correctly) or
not at all.
3.4.8
SOCIAL CONVENTIONS apply more directly to texts than to sentences. People’s
social awareness applies to occurrences, not to grammatical rule systems. The
social markedness of certain structures affects only a small portion of a total
grammar and arises only through mediation of non-sentential factors in
appropriate contexts.13
[13. The doubling of sentence subjects with noun plus pronoun, for
example, is probably due to processing strategies of the kind discussed in
V.5.8.] To approach social issues via a sentence theory, William Labov (1969)
was compelled to set up whole new provinces of rules designated “variable”
as opposed to ‘categorical.” An empirically founded linguistics will
discover, I suspect, that language rules are principle variable in accordance
with the demands of ongoing situations and with the motivations of text
producers seeking special effects.14
[24 Some unusual uses of language for special effect are found in
IV.1.17, IV.1.19, V.2.3, V.3.13, V.4.3, V.4.11, V.4.12, and VII.2.32.]
3.4.9.
PSYCHOLOGICAL FACTORS are morc relevant to texts than to sentences (cf. van Dijk
1972a: 325; Ortony 1978a: 63). In mental processing, the sentence is one
heuristic format among others for the wider purposes of communication
(O’Connell 1977), such as expressing and recovering knowledge, or pursuing a
goal. Sentence boundaries are decided late during text production and discarded
early during comprehension (Bransford & Franks 1971). A theory of sentences,
in contrast, is justified in treating as “irrelevant” such factors as
“memory limitations, distractions, shifts of attention and interest,’ and so
on (Chomsky 1965: 3f.). The wealth of experimentation purporting to explore
sentences is in principle objectionable on these grounds; however, researchers
conflated the sentence with so many other entities that their work has useful
implications for the study of texts as well
3.4.10
Texts PRESUPPOSE other texts in quite a different manner than sentences
presuppose other sentences. To utilize sentences, language users rely on
grammatical knowledge as a general, virtual system. To utilize texts, people
need experiential knowledge of specific actual occurrences (on virtual versus
actual cf. 1.4.11.6). This condition of INTERTEXTUALITY (cf. 1.4.11.6) applies
especially to summaries, replies, continuations, protocols, and parodies.
3.5
These fundamental differences between the text and the sentence m linguistic
entities have important implications for the evolution of a linguistics of the
text:
3.5.1
The confusion and conflation among entities like the sentence, the proposition,
and the speech act must be replaced by investigation of the MAPPING PROCEDURES
that control the interaction of entities on different language levels (cf.
1.2.10).
3.5.2
The search for CATEGORICAL RULES must be redirected toward the DEFAULTS and
PREFERENCES that apply with greater or lesser PROBABILITY in response to CONTEXT
(on probabilities in “performance theories, Smith 1973). Text theory cannot
state what must happen all the time but rather what is likely
to happen most of the time, given current controls.
3.5.3
Research cannot be based, nor can general conclusions be drawn, exclusively on
DEMONSTRATION SENTENCES concocted by an investigator for a particular argument.
The more convincing domain of samples is that of ACTUALLY OCCURRING TEXTS
intended to communicate (rather than to demonstrate grammatical rules). If we
are unable to find spontaneous samples in a given case, we should be cautious
about asserting the validity of our arguments. For example, sentence grammarians
have expended great research and debate upon “multiple embeddings,” which
are extremely hard to discover in real communication (cf. 11.2.27). 3.5.4. While
“much of the success” of sentence theories “is due to the strategy of
excluding unfavorable examples” (Rieser 1978: 8), the success of text
linguistics depends on a broad empirical base. We must actively seek out a
diversity of samples from all types of texts: stories, newspapers, magazines,
conversations, plays, poems, science textbooks, novels, advertisements, and many
others.
3.5.5
Text linguistics cannot accept the task of providing an abstract grammar to
generate all possible texts of a language and to exclude all non- texts. The
domain to be generated is far too vast, and continually expanding. The notion of
a “non-text” is not crucial, because the occurrence of non-texts usually
signals a refusal or inability to communicate. A more essential task for text
linguistics is rather to study the notion of TEXTUALITY as a factor arising from
communicative procedures for text utilization.
3.5.6
The models which seem most suitable for workable OPERATION in TEXT UTILIZATION
should be given the highest value as explanatory accounts. While abstract
reconstructions that somehow crank out the desired structures may be very
revealing, they should not claim to explain human language. They are at most
auxiliary and intermediary artefacts to be discarded as soon as we move closer
to a plausible model of human activities.
3.5.7
The notion of “competence” must receive a much more integrative scope than
has been customary in sentence grammars (cf. 1. 1. 17.3). We must seek to define
the abilities that make people actually competent to produce and understand
texts with consistent (though not universal) success. This kind of text theory
will be both “mentalistic” in the basic sense (cf. Fodor, Bever, &
Garrett 1974) and empirically verifiable or falsifiable.
3.5.8
Formalisms and representations must be developed that might plausibly be
interpreted as PROCESSES, not just as self-sufficient designs of unexplained
provenance (e.g., trees or formulas). A representation should suggest how the
entities in question might be BUILT, CONTROLLED, and ACCESSED (Rumelhart &
Norman 1975a: 35; J. Anderson 1976: 10; Hörmann 1976: 485; Loftus & Loftus
1976: 124; Levesque & Mylopoulos 1978: 3).
3.5.9
Whatever RULES are postulated should simultaneously embody workable PROCEDURES.
For example, the rules which build sentences ought to represent tactics that
work in real time under such normal conditions as span of memory and planning
abilities (Rumelhart 1977a: 122).
3.5.10
Our efforts must above all be devoted to INTERDISCIPLINARY CO-OPERATION.
Linguistics alone cannot provide the expertise needed to treat the
psychological, social, and computational aspects of texts in use (cf. van Dijk
1972a: 161).
3.6
I hope to make a modest beginning here toward living up to these standards. I
stress that my proposals must be tentative, pending more comprehensive research.
But I have tried to work with insights that are at least reasonable in light of
as much new research as I could assemble.
4.
TEXTUALITY
4.1 It should be noted that the general explication of the
notion of the “system” given in 1.1.6 applies not only to a language level,
but also to the entity TEXT (Hartmann 1963a: 85f.; Fowler 1977: 69). The
intersystem of a natural language such as English is composed of VIRTUAL
SYSTEMS: functional unities of elements whose potential is not yet put to use,
e.g., the repertories of sounds, grammatical forms, sentence patterns, concept
names, etc., which a particular language offers its users; in contrast to these
repertories, a text is an ACTUAL SYSTEM: a functional unity created through
processes of selection and combination among options of virtual systems
(Hartmann 1963b: 96f.; Gülich & Raible 1977: 34ff.). The evolution of a
text can therefore be termed ACTUALIZATION. This quality of occurrence, as I
have stated in 1.11f., is the essential criterion for identifying the text as
such (Hartmann 1964). It follows that the text is not simply a larger “rank”
ttan the sentence (Hasan 1978: 228), despite the views of some researchers
(e.g., Pike 1967; Heger 1976; Jones 1977). A text may be no longer than a single
word, and it may be composed of elements without sentence status (e.g. road
signs, advertisements, telegrams, and so on).
4.2
Since Saussure, linguistics has been predominantly devoted to the study of
virtual systems. Yet the knowledge of virtual systems would not be sufficient to
enable people to communicate except in a very roundabout and inefficient way.
People must know not only what options are offered, but also what options are
relevant and useable for a given situation and purpose. The virtual aspects of
mutual opposition and differentiation (following Saussure) and of
well-formedness (following Chomsky) are incomplete guidelines. I hold any
notion of competence to be incomplete which does not consider the strategies of
actualization that humans apply to virtual systems. The fact that these
strategies may lead to texts beyond the organization of the virtual systems has
been noticed in studies of poetic texts (cf. Levin 1962; Mukarovsky 1964; Thorne
1969; Beaugrande 1979e).
4.3
Actualization is a process we can explore in terms of CYBERNETIC REGULATION (cf.
Breuer 1974; Clippinger 1977). A CYBERNETIC SYSTEM possesses an internal
organization that enables it to adapt to ongoing occurrences by means of
self-regulation (cf. Klaus 1963, 1972). The main objective of the system is
STABILITY of states and operations. If the system is capable of adapting to a
variety of occurrences, it is ULTRASTABLE; if it contains several ultrastable
subsystems, it is a MULTISTABLE system (Klaus 1963: 125). The system can be
still more effective if it maintains an INTERNAL MODEL of its environment, and
if it can adapt along with the environment (a LEARNING system). All of these
attributes are assignable to a language intersystem. The functionings of virtual
systems are artificially stabilized in the abstract or synchronic viewpoint. Yet
the environment of actualization requires constant adaptation of these
subsystems according to context. As a result, the actualized text-system
reflects not only the contributing virtual systems, but appropriate
modifications and adaptations performed during the operations of actualization.
The systems remain stable if they support UTILIZATION and CONTINUITY, even
though most texts in themselves are at least partly novel and occasionally
contain greater or lesser discontinuities.
4.4
The STABILITY of the text as a cybernetic system thus depends upon the
CONTINUITY of occurrences in participating systems. This continuity is not
necessarily obvious: the stream of speech sounds or written symbols cannot
reflect all of the relations that hold the textual system together. At most, the
text is characterized by its CONNECTIVITIES, i.e. unbroken ACCESS among the
occurring elements of the participating language systems. The text users may
experience continuity as the FUZZINESS of the boundaries among the elements (cf.
III. 1.7); but the text itself can only offer connectivities. There should be
SEQUENTIAL CONNECTIVITY of GRAMMATICAL DEPENDENCIES in the SURFACE text (cf. Ch.
11). The underlying meaning should have CONCEPTUAL CONNECTIVITY, e.g. relations
of causality, time, or location (cf. Ch. III). The intentional arrangement of
the DISCOURSE ACTIONS within texts (cf. III.4.26) should reflect a PLANNING
connectivity, so that each component utterance is RELEVANT to some interactive
or communicative plan, such as advising, requesting, agreeing, or just
maintaining social solidarity (cf. VI.4; VIII.1). The context determines how
many actual occurrences are needed for connectivity to prevail. In highly
determinate contexts, people economize by omitting or compacting the actual
occurrences in the surface structure of expression.
4.5
Illustrations of the regulatory operations I have outlined are not hard to find
in regard to texts:
4.5.1
The compacting of surface structure in determinate contexts can be performed by
the use of pro-forms and ellipsis (cf. V.4 and V.6).
4.5.2
Decisions about the organization of a conceptual configuration elicit follow-up
decisions about the organization of surface structure and vice versa (cf.
III.4.16; VII.2.10ff.)
4.5.3
In the presence of ambiguities or disturbances, people can make intensified use
of other cues to maintain textuality (see e.g. 11.2.37; V.4.1 1).
4.5.4
When elements of a presented text are forgotten, the textual system in mental
storage adjusts by compacting, rearranging, or reconstructing the remainder (see
section VII.3).
4.5.5
Discrepancies and discontinuities do not normally cause a breakdown in
communication, but only elicit regulatory operations from the hearer or reader
(see I.6.9; IV.1.12; VIII.2.42ff.).
4.6 Most important of all is the regulatory nature of communicative systems among individual participants. Every person’s knowledge and experience are in some ways unique, yet people normally communicate without difficulty. If a given person fails to utilize some language subsystem in the conventional way, regulatory occurrences generally become necessary: explaining, restating, correcting misunderstandings, precluding alternative readings, and even apologizing. People’s actions and utterances are not governed by laws or categorical rules; but people must respect the operations of a system if they intend to use it effectively. Individual misuses are rare precisely because they elicit regulatory occurrences that damage efficiency if repeated very often.
4.7 Many texts are manifestly able to survive and be utilized long after their original contexts have been lost. If the virtual language systems undergo changes in the intervening time, readers need some mediation, such as the training needed today to read Old or Middle English. However, if the virtual systems remain generally stable, utilization is unproblematic. Texts are SELF-CONTEXTUALIZING because the actualization processes of writers and readers are geared toward continuity and regulation (cf. Halliday, Mclntosh, & Strevens 1965: 246; R. Anderson 1977: 242). The higher the quality of a text, the greater its potential for later utilization: the decisions and selections made in production are especially well designed in enduring works (Winograd 1977a. 69). This factor accounts for the endurance of literary and poetic texts over other types (cf. VII.2.37ff.).
4.8
The possibility that different hearers or readers might make different uses of
the same text is by no means unproblematic, as the lively debates over the role
of the reader in literary theory attest (cf. Warning [ed.] 1975; Beaugrande
1988). The stability of the text is derivative from the stability of
participating virtual systems of communication, and the regulatory principles of
actualization-a kind of “meta-stability” (E. D. Hirsch, personal
communication). Linguistic discussions have often missed these considerations by
dwelling extensively upon potential ambiguities or alternatives allowed by
virtual systems, and giving little heed to the fact that real utterances are
seldom misunderstood. The picture of language processing that emerges from those
discussions is one of the language user floating in a sea of alternative
readings and structures whose management in any reasonable time span seems
miraculous. Thus, Chomsky (1975: 77) concludes: “the study of the capacity to
use these structures and the exercise of this capacity, however, still seems to
elude our understanding.”
4.9
Transformational grammar is a pre-eminently virtual system that undertakes to
state which sentences are categorically possible without regard for their
occurrence. Linguists’ samples are to some extent pseudo- occurrences, unless
taken from spontaneously produced texts of non- linguists. Yet a grammar of
pseudo-occurrences is a curious construction for a science, and its verification
already a grave problem (cf. 1.1.16ff.). Surely the enumeration of all possible
sentences becomes a performance issue after the CORE of the grammar is
systemized (cf. Grimes 1975: 198). People’s competence is above all their
limited, operational set of strategies for building and understanding sentences
or texts that are likely to occur because they make sense and are useful in
getting things done.
4.10
It is not surprising that linguists initially hoped to treat texts as virtual
systems or system elements. Harris’s (1952) attempt to uncover the
distributional rules for texts suggests the assumption that virtual and actual
systems were convergent. A pilot project largely inspired by transformational
grammar was devoted to creating a rule apparatus to generate or derive a text by
Bertolt Brecht (van Dijk, Ihwe, Petöfi, & Rieser 1972; see the debate
between Ihwe & Rieser 1972 and Kummer 1972b, 1972e over the outcome). Thomas
Ballmer (1975: 259) sees texts as nothing more than “well-formed sequences of
morphemes” which can be treated by enlarging sentence grammar with
“punctuation morphemes.” These and similar experiments are all subject to
the same principled objections: (1) they provide no plausible model of human
activities; (2) they are operationally unworkable for any significantly large
corpus of texts; and (3) they do not deal realistically with such issues as
ungrammatical texts, better or worse style, interestingness, informativity, and
communicative interaction.
4.11
I propose the following standards of TEXTUALITY to be the legitimate basis of
the actualization and utilization of texts:
4.11.1
COHESION subsumes the procedures whereby SURFACE elements appear as progressive
occurrences such that their SEQUENTIAL CONNECTIVITY is maintained and made
recoverable. The means of cohesion include the grammatical formatting of
phrases, clauses, and sentences (see Chapter 11), and such devices as
recurrence, pro-forms and articles, co-reference, ellipsis, and junction (see
Chapter V).
4.11.2
COHERENCE subsumes the procedures whereby elements of KNOWLEDGE are activated
such that their CONCEPTUAL CONNECTIVITY is maintained and made recoverable. The
means of coherence include: (1) logical relations such as causality and class
inclusion; (2) knowledge of how events, actions, objects, and situations are
organized; and (3) the striving for continuity in human experience. Cohesion is
upheld by continual interaction of TEXT-PRESENTED KNOWLEDGE with PRIOR KNOWLEDGE
OF THE WORLD (cf. VII.3.29ff.).
4.11.3
INTENTIONALITY subsumes the text producer’s attitude that a given language
configuration is INTENDED to be a cohesive and coherent text; and that such a
text is an INSTRUMENT in following a PLAN toward a GOAL (cf. VI.4). There is a
variable range of TOLERANCE where intentionality remains in effect even when the
full standards of cohesion and coherence are not met, and when the plan does not
actually load to the desired goal. This tolerance is a factor of systemic
regulation (1.4.3f.) mediating between language strategies at large and the
exigencies of ongoing contexts.
4.11.4
ACCEPTABILITY subsumes the text receiver’s attitude that a language
configuration should be ACCEPTED as a cohesive and coherent text. Acceptability
also has a tolerance range for cases where context brings disturbances, or where
the receiver does not share the producer’s goals (cf. 11.2.37f.).
4.11.5
SITUATIONALITY subsumes the factors that make a text relevant to a current or
recoverable SITUATION. The text figures as an ACTION that can both monitor and
change a situation (cf. VI.4.2ff.). There may be only slight mediation toward
the situation, as in face-to-face communication about directly apperceivable
events; or substantial mediation, as in reading an old text of literary nature
about events in an alternative world (e.g. Gilgamesh or The Odyssey)
(cf. VII. 1.8.4). The scope of situationality always implies the roles of at
least two communicative participants, but they may not enter the focus of
attention as persons.
4.11.6
INFORMATIVITY is the factor of the relative UNCERTAINTY about textual
occurrences or occurrences within a textual world as opposed to possible
alternatives. Informativity is high if the alternatives are numerous and if an
improbable alternative is actually selected. However, every text has at least
the minimal informativity in which its occurrences are opposed to
non-occurrences (cf. IV. 1.8). I argue in Chapter IV that a medium degree of
informativity is maintained in communication by means of regulating extreme
degrees.
4.11.7
INTERTEXTUALITY subsumes the relationships between a given text and other
relevant texts encountered in prior experience, with or without mediation. A
reply in conversation (cf. VIII.1) or a recall protocol of a text just read (cf.
VII.3) illustrate intertextuality with very little mediation. More extensive
mediation obtains when replies or criticisms are directed to texts written down
at some earlier time. Intertextuality is the major factor in the establishment
of TEXT TYPES (cf. VIII.1), where expectations are formed for whole classes of
language occurrences.
4.12
These standards are of course not new, but their treatment hitherto has been
sporadic and diffuse. Cohesion and coherence, for example, have often been
conflated, due perhaps to the widespread confusion regarding the nature of the
sentence (cf. 1.3.1ff.) (but cf. Widdowson 1973). The notions of cohesion and
coherence can be pursued in such works as Halliday (1964); Crymes (1968); Harweg
(1968a); Hasan (1968); Palek (1968); Bellert (1970); van Dijk (1972a); Grimes
(1975); Hobbs (1976, 1979); Halliday & Hasan (1976); Bullwinkle (1977);
Jones (1977); Reichman (1978); Webber (1978). On intentionality, compare
Wunderlich (1971); Dressler (1972a); Bruce (1975); van Dijk (1977a); Schlesinger
(1977); Cohen (1978); Allen (1979) (more literature on plans and goals is given
in chapter VI). Concerning acceptability, consult Quirk & Svartvik (1966),
and Greenbaum (ed.) (1977). On situationality, the work of Halliday (e.g. 1977)
and the ethnography of communication (e.g. Gumperz & Hymes [eds.] 1972) are
relevant. For some outlooks on intertextuality, consider Kristeva (1968) and
Quirk (1978). Regarding informativity, little is available except on “given’
and “new” knowledge in sentences as reviewed in section IV. 3; but cf.
Shannon (1 95 1); Weitner (1964); Grimes (1975); Groeben (1978); Beaugrande
(1978b, 1979e). All seven criteria of textuality are discussed in turn in
Beaugrande & Dressler (1981).
4.13
Of these seven criteria, two seem prominently text-oriented (cohesion and
coherence), two prominently psychological (intentionality and acceptability),
two prominently social (situationality and intertextuality), and the last,
computational (informativity). But close investigation shows that none of the
criteria can be appreciated without considering all four factors: language,
mind, society, and processing. Again, the pressing need for interdisciplinary
research stands forth. These criteria of textuality figure as CONSTITUTIVE
principles in the sense of Searle (1969. 33ff.): whether or not something can be
considered a text depends on whether these criteria are upheld. There must also
be REGULATIVE principles in the sense of Searle which distinguish the quality of
a sample already admitted as a text. I surmise that this regulative function is
exercised by the criteria of design I shall propose.
4.14 While all texts must possess these standards of
textuality, there are differences in the DESIGN of their actualization. We must
therefore define and investigate DESIGN CRITERIA such as the following (for
discussions and illustrations see III.3.5; IV.1.6; IV.4.12., VII.2.37;
VIII.2.19). The EFFICIENCY of a text results from its utilization in
communication with the greatest returns for the least effort, so that PROCESSING
EASE is promoted. The EFFECTIVENESS of the text depends upon its intensity of
impact on text receivers, promoting PROCESSING DEPTH, and upon its contribution
toward the producer’s goal, constituting the RELEVANCE of text materials to
steps in a plan. The APPROPRIATENESS of a text depends on the proportionality
between the demands of a communicative situation and the degree to which
standards of textuality are upheld. These design criteria are, I believe, much
more vital to language users “competence” than the famous distinction
between sentences and non-sentences, or a parallel distinction between texts and
non-texts. Normally, the production of non-texts signals a refusal or inability
to communicate at all (cf. 1.3.4.4; IV.1.23.2; V.4.12). Thus, the absence of
cohesion, coherence, intentionality, informativity, etc. is comparatively rare;
but texts may often be inefficient, ineffective, or inappropriate. We must study
not only how language structures can be built and analysed, but also how they
can be evaluated.
5.
TEXTUAL COMPETENCE
5.1
To deal with manifestations and data of any sort, a science must differentiate
between the essential, regular, and relevant aspects and the non-essential,
idiosyncratic, or irrelevant ones. For example, phonology studies systems of
sounds by discounting such factors as the voice quality, age, sex, or
personality of speakers; otherwise, no two sound patterns of the same utterance
would ever be exactly identical. To establish a theory of sentence grammar,
Chomsky (1965) eliminated such factors as memory limitations, changes of plan
while speaking, and errors.
5.2
The distinction of competence and performance along the lines of sentence
grammar has come increasingly under fire in recent years. Walter Kintsch (1974:
3) adjudges the distinction as “merely an excuse for both the linguist and the
psychologist to justify the neglect of each other’s findings.” Werner Kummer
(1975: 163) discards the distinction as inherently tied to a language model
incapable of integration into a theory of action. Other researchers retain the
distinction while calling for a new orientation toward COMMUNICATIVE COMPETENCE
(cf. Wunderlich 1971; Habermas 1971; Hymes 1972; Schmidt 1973).
5.3
I also hold the distinction to be valuable if not indispensable:
5.3.1
Whether we look at the physical properties of a stream of speech or at the wider
constellation of a communicative situation, we are forced to admit that elements
we consider equivalent in their systemic functions are superficially
different in minor though discoverable ways. To communicate at all, people
must distribute their attention and resources selectively toward relevant
aspects, while disattending the rest. The linguist is certainly justified in
emulating this selectivity without which language could never be subjected to
identification, generalization, description, and the other tasks enumerated in
1.1.8
5.3.2
To deal adequately with any representative sampling of texts in a language, we
must specify a reasonably limited set of strategies and procedures that
apply to very diverse manifestations. Competence must cover these shared
abilities even though some manifestations may be impaired by restrictions of
time, resources, attention, knowledge, or experience on the part of an
individual language user.
5.3.3
Language activities are frequently creative.
Many texts convey informativity by virtue of their producer’s modifications
upon the normal or expected organization of texts (cf. Beaugrande 1979e). If we
undertook to incorporate every manifestation of creativity into the same
framework as conventional procedures, we would misrepresent many issues. A
grammar that could produce every creative configuration would eventually be
powerful enough to produce every conceivable configuration, thus attaining zero-
organization and accounting for nothing at all.15
[15. In a system with zero organization, no predictions can be
made about any occurrences or regularities.]
5.4
I would conclude that manifestations reflect competence, but they need
not embody it (cf. I.4.14). The competence/ performance distinction
should be retained as the opposition between STRATEGIES (procedures applied and
held to be useable most of the time) and APPLICATIONS (the detailed events of
communicative situations, including disturbances or failures). We should explore
the effects that arise when strategies do not work; speech errors, for example,
provide valuable evidence regarding mental operations (cf. Fromkin [ed.] 1973;
Goodman & Burke 1973; examples in VII.3.14ff.). But it would seem odd to
conflate competence with performance by postulating strategies that are designed
to produce errors and failures.
5.5
I would not define competence solely as the ability to distinguish between texts
and non-texts.16 [16. As Schank and Wilensky (1977: 142) point
out, the distinction of “grammatical vs. ungrammatical” is unrealistic,
because “people don’t go around trying to distinguish English from
gibberish.”] Except in the presence of special signals, people probably make
the DEFAULT ASSUMPTION that language presentations are texts. The notion of
TEXTUAL COMPETENCE (van Dijk 1972a: 204) might rather be required to subsume the
following set of KNOWLEDGE and PROCEDURES:
5.5.1
knowledge of the repertories of OPTIONS in virtual systems of language;
5.5.2
knowledge of systemic CONSTRAINTS on the selection and combination of options;
5.5.3
knowledge of the BELIEFS, KNOWLEDGE, and EXPECTATIONS shared by the
communicative group or society about the “real world”; [171 view the ‘real
world” not as some irrefutably given set of objects, but rather as the
socially accepted model of whatever objects are there (cf. IV. 1.21.3).17
[17. 17 I view the ‘real world” not as some irrefutably given set of
objects, but rather as the socially accepted model of whatever objects are there
(cf. IV. 1.21.3).]
5.5.4
knowledge of TEXT TYPES;
5.5.5.
procedures for UTILIZING virtual systems during ACTUALIZATION;
5.5.6
procedures for PRODUCING texts;
5.5.7
procedures for RECEIVING texts;
5.5.8
procedures for maintaining TEXTUALITY;
5.5.9
procedures for regulating INFORMATIVITY;
5.5.10
procedures for optimizing DESIGN CRITERIA (efficiency, effectiveness, and
appropriateness);
5.5.11
procedures for re-utilizing text-acquired knowledge from mental storage in tasks
like RECALLING, REPORTING, SUMMARIZING, or EVALUATING;
5.5.12
procedures for MONITORING and MANAGING SITUATIONS by using texts;
5.5.13
procedures for building, implementing, and revising PLANS toward GOALS;
5.5.14
procedures for PREDICTING the activities of other participants in communication
and REGULATING one’s discourse actions accordingly;
5.5.15
procedures for maintaining communication despite DISCREPANCIES, DISCONTINUITIES,
AMBIGUITIES, or NON-EXPECTED occurrences.
5.6
I would surmise that INTELLIGENCE can be defined as the decoupling of these
skills and processes from any particular task at hand. It is the capacity to
operate on a higher plane and to recognize and perform any given task as an
instantiation of a general operation type; and to treat any given data in terms
of a general data type. I shall accordingly propose that textual communication
functions on a high plane: syntax, meaning, information, and planning are
processed in terms of high-plane typologies of occurrences and relationships
(cf. II.2.15ff.; III.4.3ff.; IV.1.6ff.; IV.3.17ff.; V.1.4ff.; VI.I.Iff.;
VI.4.14; VII.I.7; VII.2.8ff.; VII.3.15ff.; VII.3.29ff.; VIII.2.8ff.;
VIII.2.21ff.; IX.1.4ff.). I suspect that the continuing failure of linguists to
solve or explain many major issues of language communication has been due to
adopting an unduly low-level perspective (analysing the meanings of individual
words or the exact surface formats of specific sentences, and so on) (cf. IX.8).
6.
TEXT UTILIZATION AS MODEL-BUILDING
6.1
The activities involved in the production and comprehension of a text can be
explored in terms of MODEL-BUILDING. The participants in communication can be
said to be BUILDING A TEXT-WORLD MODEL (cf. the notions of “world’ or
“model” in Petöfi & Rieser 1974; Petöfi 1975a; Schank et al. 1975;
Collins, Brown, & Larkin 1977; Fahlman 1977; Goldman, Balzer, & Wile
1977; Reichman 1978; Rubin 1978; Webber 1978; Petöfi 1979). The TEXTUAL WORLD
is the cognitive correlate of the knowledge conveyed and activated by a text in
use. As such, it is in fact only present in the minds of language users. Hence,
we must approach the problem via the MODELS of textual worlds as composed of
CONCEPTS and RELATIONS in a KNOWLEDGE SPACE (cf. Ch. Ill). The text-world model
is viewed as embedded in a SITUATION MODEL (cf. Clark & Clark 1977: 72;
Grosz 1977: 6). The situation model is kept in alignment with the PLANS and
GOALS of the participants (in this sense, a goal is a model of a desired future
situation — cf. VI.4.4). The text producer can maintain a model of the text
receivers and their knowledge (cf. Bruce 1975: 5; Goldman 1975: 346; Bernstein
& Pike 1977: 3; Winograd 1977a: 69, Cohen 1978: 16; McCalla 1978a: 19;
Carbonell Jr. 1978b; Rubin 1978b: 136; Allen 1979: 6). We could go on to
postulate the receivers’ model of the producer’s model of them, and the
latter’s model of their model, and so on (see Clark & Marshall 1978). But
there is probably a THRESHOLD OF TERMINATION where people in communication do
not bother to run through all these models inside models.
6.2
An integrated approach entitled the TEXT-STRUCTURE/ WORLD- STRUCTURE THEORY has
been set forth by János S. Petöfi and associates (Petöfi 1975a, 1975 b,
1978a, 1978 b, 1979; Biasci & Fritsche [eds.] 1978). The basic postulate in
the theory is that there are regular correspondences between the structure of a
text and the structure of the “world” a text evokes. Petöfi (1978a: 44f.)
notes that there are two outlooks on the development of such an integrated
theory: One can either set out from an existing apparatus (with its limitations
but well known scope) and try to modify it to the extent required by the object
under investigation; or one can start with what is required for the description
of the object and try to devise an apparatus accordingly. Petöfi has proceeded
by working with the “existing apparatus” of formal logics. But he realizes
the need for substantial modifications, e.g.: “the rule systems of logical
syntaxes in use so far are not suitable for the description of natural
languages, because the logical formulae assigned by them to natural language
utterances are only capable of representing a part of the syntactic information
found in natural language utterances” (Petöfi 1978a: 40). The latest version
(Petöfi 1979) foresees an elaborate network of components such as lexicon,
canonic language, natural language, description, interpretation, formation,
composition, transformation, 18 [18. Petöfi’s
‘transformations’ are not like those of usual sentence grammars, because
they convert structures into structures of different systemic types (cf.
II.1.6).] and representation. The canonic language is made flexible via
expansion of the object domain and yet is still translatable into first-order
predicate calculus. The model-building function is handled by the interpretation
component. An unusual feature not found in conventional logics for natural
language is Petöfi’s attempt to deal with the apperception and description of
language sounds.
6.3
One difficult question concerns the nature of a WORLD, i.e., the totality of
data given in some context. In the tradition of Carnap and Kripke, the LOGICAL
WORLD is ATOMISTIC (Cresswell 1973:38; cf. Hughes & Cresswell 1968). The
atomism arises from the DISCRETENESS of objects and functions as required by the
formatting and proof techniques of the logic. Hence, content appears as MODULAR
and insensitive to many kinds of context. Max Cresswell (personal communication)
tells me that attempts are under way to overcome atomism by plotting logical
worlds close together on a CONTINUUM (see also Eikmeyer & Rieser 1978). The
work on FUZZY SETS by Lotfi Zadeh and others allows the introduction of
indistinct or probabilistic boundaries among entities of meaning. These
important advances do not in themselves stipulate what the human processes of
utilizing knowledge ought to look like. Hence, I follow the other outlook cited
by Petöfi and explore what might be required for representations that could be
developed in the future.
6.4
A TEXTUAL WORLD plainly has great potential for CONTINUITY. The spaces between
text-presented concepts and relations can be filled in or enriched with a wide
range of COMMONSENSE KNOWLEDGE about how events, actions, objects, and
situations are organized. Three factors should be cited here. SPREADING
ACTIVATION occurs when the material activated by a text contacts associated
material already stored in the minds of text users (e.g., for building up a
scene from a few details mentioned in the text) (cf. III.3.24). INFERENCING is
done whenever GAPS are noticed among points in a knowledge space (e.g., for
solving a crime in a detective story) (cf. Rieger 1974, 1975, 1976; Clark 1977;
Collins, Brown, & Larkin 1977; Warren, Nicholas, & Trabasso 1979; cf.
III.4.29ff.)19 UPDATING
changes the textual world regarding what is true at any moment as the course of
events affects the situation (cf. Sacerdoti 1977: 15; Winston 1977: 386). The
extent to which these three processes are actually carried out may vary among
individual languages users; empirical testing will, I believe, show a THRESHOLD
OF TERMINATION where continuity is considered satisfactory and these processes
stop. In any case, these processes make it unnecessary for a text producer to
explicitly state all the material needed for coherence.
6.5 Two well-known approaches to model-building correspond to
the two approaches cited in the Petöfi quote in I.6.2. The INDUCTIVE approach
works by reacting to and generalizing from observations and experience; the
DEDUCTIVE approach entails a prior stipulation of what some domain ought to be
like. This distinction applies to building text-world models also, especially
from the hearer/ reader’s standpoint. People notice and classify the incoming
presentation as BOTTOM-UP input; on the other hand, they steadily form and test
hypotheses about what will occur or be stated, applying 191 further distinguish
between spreading activation vs. inferencing in TOP-DOWN input (on top-down vs.
bottom-up, see R. Bobrow & Brown 1975; Bobrow & Norman 1975; Brown &
Burton 1975; Collins, Brown, & Larkin 1977). In this view, the task of
understanding is one of integrating presented knowledge into stored knowledge
(Kintsch 1974: I 1; cf. Ausubel 1963 on “subsumption’).
6.6
To decide what knowledge should be applied, cognitive processing operates by
PATTERN-MATCHING (cf. Colby & Parkinson 1974; D. Bobrow 1975; Rieger 1975,
1976; Rumelhart 1975, 1977a; Kuipers 1975; J. Anderson 1976; Kintsch 1977a;
Winston 1977; Bobrow & Winograd 1977; Hayes 1977; Pavlidis 1977; Havens
1978). A perfect match is not required, but only a reasonably good fit (cf.
Rieger 1975: 277; Woods 1975a: 36). For efficiency, it is desirable to match the
largest possible pattern and thus treat the greatest amount of input at once (Rieger
1975: 157).
6.7
The best means for representing the procedures of model-building and pattern
matching in textual communication is, in my view, GENERAL PROBLEM SOLVING (cf.
Newell & Simon 1972; Winston 1977).20 [20.
The term “general” was used to suggest that the first program written by
Allen Newell, Herbert Simon, and Cliff Shaw in 1957 was divided into “a
task-independent part of the system containing general problem solving
mechanisms” and “a part of the system containing knowledge of the task
environment” (Newell & Simon 1972:414; cf. IX.1.7). The use I make of
general problem-solving here is rather different from that envisioned by these
scholars, though, I hope, in line with their general theory.]
A PROBLEM can be defined as a state from which the pathway to the
successor state has a noticeable probability for FAILURE: the pathway is not
traversed because either the pathway or the successor state is wrong. The
PROBLEM-SOLVER is a PLANNER, which has to SEARCH the PROBLEM SPACE in order to
connect the current state with the intended successor state. A SERIOUS PROBLEM
is present when the probabilities for FAILURE are higher than those for SUCCESS.
A BLOCK is present if the planner cannot advance at all; it is then necessary to
discard the pathway and go to a point where new progress can be made. As we can
see, problem solving depends chiefly on techniques of SEARCH, of which at least
three kinds should be mentioned (cf. Lenat 1977: 1099f; Winston 1977: 90ff.,
130ff.):
6.7.1 In MEANS-END ANALYSIS, the processor focuses on the major differences between the first point (the INITIAL STATE) and the final point (the GOAL STATE). All operations are directed to reducing the differences between these two states. In forward progression, means-end analysis resembles depth-first search as explained in the following s