Poetics 14 (1985) 551-574
North-Holland
551
O N T H E R O L E O F C O N V E N T I O N S IN U N D E R S T A N D I N G
LITERARY TEXTS
Dietrich M E U T S C H and Siegfried J. S C H M I D T *
On the basis of the empirical science of literature and new results of text processing research five
hypotheses are developed. They concern the role of the polyvalence convention while understanding literary texts. The results of the pilot study indicate the existence of polyvalent discourse
processes, but they also point to the necessity of further research to look for the influence of
different literary texts on cognition.
I. I n t r o d u c t i o n
In what follows we report the results of a pilot study concerning problems of
understanding literary texts. The research work was carried out at the University of Siegen from 1984 to 1985 by the two authors, assisted by Henrike Alfes.
In our project special attention was paid to explicitly displaying the connection between the presupposed theories and models and the hypotheses derived
from them; the selection of methods, and the collection and interpretation of
the data.
The theoretical basis presupposed in our research project consists of two
domains:
- a theory (or, to put it more modestly: a model) of understanding (Meutsch
1986a, Schmidt 1985, Viehoff and Schmidt 1985), and
- an empirical theory of literature (Schmidt 1980, 1982, Hauptmeier and
Schmidt 1985).
A special task of the latter theory was to provide us with the key concept of
our project: the concept of polyvalence. Our project aimed at three goals: to
attain a better explanation of the understanding of literature texts, to find an
empirical confirmation of the existence and the effects of the polyvalence
convention as predicted by the empirical theory of literature and to prepare a
complex and new research project (Schmidt 1985).
* Mailing address: S.J. Schmidt, Universit~it Siegen, Fachbereich Sprach- und Literaturwissenschaften, Adolf Reichwein-Strasse, Postfach 210209, D-5900 Siegen 21, FRG.
0304-422X/86/$3.50 © 1986, Elsevier Science Publishers B.V. (North-Holland)
552
D. Meutsch, S.J. Schmidt / Understanding literary texts
2. Towards a model of understanding
2.1
Recent research in reading and comprehension has revealed some interesting
points of convergence. Psycholinguistic as well as cognitive psychological
models of understanding assume that text understanding can be described in
terms of text processing which is theoretically modelled as an interaction of
two domains of information: those gathered from the text (= bottom-upoperations) and those derived from the cognitive domain (= top-downoperations; cf. Mandler and Johnson 1980, Rumelhart 1980, van Dijk and
Kintsch 1983, Just and Carpenter 1980). Text processing is regarded as a
process which has to be characterized by means of dynamic categories. In
addition to relations between text and cognitive domain (in terms of memory,
knowledge, strategies, etc.), recent research tries to take into account social
and contextual conditions of text processing, the role of acting knowledge, the
influence of non-verbal elements, and the like (e.g. Mandl and Schnotz 1985,
Mandl, Stein and Trabasso 1984).
Text processing is regarded as a sequential, cyclical, hierarchically ordered,
strategy-guided and context-bound process aiming at a coherent organisation
of text data and knowledge structures. The strategies operating on this process
are usually labeled as micro-, macro-, and superstrategies (cf. van Dijk and
Kintsch 1983). The knowledge invested in text processing relates to four
dimensions: text, world, story structures, and structures of story telling.
After the so-called 'cognitive turn' in psychology more and more evidence
has been supplied for the hypothesis that text processing is an active and
constructive process in the cognitive domain of understanding subjects. (For a
review see Groeben 1982a, Schmidt 1985, Meutsch 1986a, Ballstaedt, Mandl,
Schnotz and Tergan 1981, Mandl and Schnotz 1985, Rickheit and Strohner
1985, v. Dijk 1985, Flammer and Kintsch 1982.)
2.2
In our view the seemingly evident and consensual assumption that understanding equals operations of information-processing is problematic. The concept of
information underlying this idea has been simply adopted from information
theory, taking for granted that information is something that can be exchanged, transmitted, etc. In contrast, biologists and psychologists like
Maturana (1982), Johnson-Laird (1983), or von Glasersfeld (1986) hold the
view that information is constructed inside the cognitive domain of subjects.
In verbal interactions (communication) it cannot simply be exchanged but
must be construed analogously in the respective cognitive domains of the
interlocutors. As a result of these (empirically backed) ideas, we prefer to
D. Meutsch, S.J. Schmidt / Understanding literary texts
553
model cognition as a process of information-construction instead of information-processing (cf. H/~rmann 1976, and Schnotz 1985).
The same type of argumentation concerns the intuitively plausible idea of
an interaction between text and receiver in understanding processes, where the
text is regarded as active a factor as the recipient. Not only neurobiologists but
also psychologists have opposed this idea with important arguments.
2.3
We cannot go into the details of constructivist theories of language, meaning,
and communication. 2 The most important argument we accept from these
theories reads as follows. Due to the operational closedness of our nervous
system, all instances of 'reality', 'sense', 'information', 'value', etc. are bound
to activities in the cognitive domain of subjects (whose social acting conditions
are equally respected as the biological ones). Accordingly, texts as meaningful
instances cannot be described apart from their being construed as meaningful
instances in the cognitive domain of subjects; i.e. as meaningful instances they
do not exist outside these cognitive domains.
In a constructivist approach understanding is hence theoretically modelled
as a relation between a material (graphematic or acoustic) text-basis ( = TEXT)
and the cognitive domain of a socialized subject in an actual social and
situational acting context. The result of performing this relation is what we
call a K O M M U N I K A T , 3 i.e. an internal mental representation of a TEXT.
2.4
On the basis of these general hypotheses, 'understanding' can be conceived of
theoretically as a process of Kommunikat-construction. Making use of the
results of hitherto existing comprehension research, 4 this process can be
characterized by the following features which we regard as being essential:
(a) The process of Kommunikat-construction (abbreviated as PKC) is integrated into encompassing practical activities of the actor (Anderson 1985).
1 Cf. Kintsch (1978: 66): 'For the psychologist, text bases, and hence meaning, are not real-world
objects but simply the outcome of certain psychological processes. When we read a text, the only
thing outside the reader's mind are some letter shapes on a page: the words that are communicated by these visual objects, the phrases and sentences in which these words are organized, and
their meaning, are the result of complex hierarchical psychological processes taking place in the
reader's mind. In particular the meaning of the text that the reader constructs from the actual
physical stimulus, the situational context, and his world knowledge, is not something that resides
in the text, but a response on the part of the rader to the text'. See also McCormick-Leighty
(1985).
2 For further details see Schmidt (1983).
3 We make use of the German words in order to avoid equivocations with c o m m o n English terms.
4 See the report on the state of research in Hoppe-Graff (1984).
554
D. Meut'sch, S.J. Schmidt / Understanding literary texts
(b) PKC can be theoretically modelled as a holistic process which hierarchically organizes and integrates component processes (cf. van Dijk and Kintsch
1983, Meyer 1984, Schnotz 1985) that differ with regard to their states of
consciousness and awareness (van Dijk and Kintsch 1983, Viehoff and Meutsch
1985). The process as a whole can analytically be divided into three interrelated domains: a rational domain of intentional informational self-orientation
and self-reflection of the subject; an emotional domain, and a domain of
evaluative checks of the practical relevance of PKC (and its results) for the
subject (cf. Spiro 1982, D~Srner 1983, HiSrmann 1983, Klix 1980, Kluwe
1979). 5
(c) PKC can be described as an intersection of knowledge acquired and
developed in processes of socialization and actual activities (cf. Leont'ev 1979,
Neisser 1976, D6rner 1976, Mandl and Schnotz 1985, Tergan 1984). Here we
have to take into account that knowledge is always institutionally determined
(cf. Verdaasdonk 1982).
(d) PKC is regarded as a process which is governed by cognitive schemata.
Following Anochin (1976), Galanter, Miller and Pribram (1960) and Hacker
(1978, 1980), we assume that anticipatory cognitive schemata (see Hacker's
'operative systems of representation', 1978, and von Cranach, Kalbermatten,
Indermtihle and Gugler 1980) bring about the program of an action (Stadler,
Schwab and Wehner 1979) and achieve its perceptual feedback.
(e) According to HiSrmann's postulate of 'sense constancy' (1976), PKC is
intentionally structured by the cognizing subject and basically oriented towards coherence. This postulate is also relevant to literary reception (cf.
Schmidt 1984). Specific aesthetical knowledge and values constrain the types
and degrees of coherence which are realized (cf. Viehoff 1976,1981, Harker
1981). Here we have to make differences between micro-structural types of
coherence (as can be found in processing metaphors, see Bates and McWhinney 1979), super-structural preferences (cf. van Dijk 1979, van Dijk and
Kintsch 1983), or individual or idiosyncratic reading strategies (cf. Meutsch
1986a; Meutsch and Viehoff 1985).
(f) PKC depends on biological conditions as well as on those cognitive acting
conditions which the subject acquires through his/her process of socialization
(Viehoff 1982).
(g) PKC is governed by concept-dependent comprehension strategies. Different strategies have been investigated in recent research. Vipond and Hunt
(1984) distinguish between point-driven, story-driven, and information-driven
readings, thereby illustrating differences between reading strategies in relation
to socio-functional constraints acting on individual readers. Van Dijk and
5 Perhaps the rational domain is bound to the structure of information construction whereas the
emotional and relevance domains provide the dynamics of PKC, while all three are subject to the
overriding principle of all living systems, i.e. the maintaining of its autopoiesis.
D. Meutsch, S.J. Schmidt / Understanding literary texts
555
Kintsch (1983) claim the efficiency of propositional, local coherence, macroand schematic strategies, thereby emphasizing text-oriented understanding
without regarding social influences (see also Meyer 1984, Meyer and Rice
1982).
(h) Special features of text organisation and text presentation trigger special
reading strategies which can be elaborated according to goals and contexts. In
this process cognitive schemata can be produced or modified (cf. Meutsch
1984, 1986a, Graesser et al. 1979, Graesser 1981).
(i) The course of PKC is characterized by back- and forth-oriented operations (cf. Crothers 1979, Clark 1977, Warren, Nicholas and Trabasso 1979).
Back-oriented operations normally contribute to an enlargement of coherence,
whereas forth-oriented ones modify the reader's expectations of possible
coherence (cf. Schank and Lebowitz 1980 on the 'predictivity of understanding').
(j) PKC is determined by expected functions of the reading process (i.e. by
exigencies, expectations, and conditions of understanding).
(k) PKC is governed by conventions. We distinguish between general conventions (of the Gricean and Searlean type), and special conventions peculiar
to special discourses. The polyvalence convention is supposed to belong to the
set of these special conventions in our society (Schmidt 1982).
(1) PKC is a constructive, subject-dependent process. Intersubjectivity in
PKCs of different subjects is completely bound to computable processes of
socialization and culturalisation as well as to a mutual acceptance of conventions by interacting subjects (cf. Kintsch and Greene 1978) and to a comparable interpretation of the respective communication situation.
3. Some essentials of the presupposed theory of literature
3.1
In numerous publications over a decade S.J. Schmidt and his NIKOL-research
group have put forward a lot of arguments why literary scholars should
transform their research domain and proceed from the analysis of literary
texts to the description, explanation, and modification of literary systems.
As has been explicitly developed in Schmidt (1982) and Hauptmeier and
Schmidt (1985), the smallest units under investigation in literary studies
should not be isolated (and implicitly ontologized) literary texts, but such
text-actor-context-syndromes in which phenomena are focussed which the
actors deem literary. These actions are called literary actions and subsumed
under four basic acting roles, i.e. the production, mediation or distribution,
reception, and post-processing of literary texts. Concatenations of literary
556
D. Meut'sch, S.J. Schrnidt / Understanding literary texts
actions are called literary processes. Nettings of literary processes constitute
what is called a literary system.
From a macrosociological point of view literary systems are regarded as
component systems of a society, which in turn is theoretically modelled as a
system consisting of component systems. As a component system of society, a
literary system is in constant exchange and interaction with other component
systems of a society like politics, economy, education, religion, etc. According
to system-theoretical approaches in sociology, social systems normally have to
be characterized with regard to their structure, their function, and their
inside-outside-differentiation.
For the purposes of this paper we can restrict ourselves to the last point.
We suppose that the demarcation line between the literary system and all
other component systems of modern societies is drawn by two conventions
which govern the ways literary phenomena are produced, mediated, received,
and post-processed in modern literary systems.
3.2
It should be emphasized that the two conventions which specify literary
actions and literary processes work on (or dominate) two other conventions
which, on the opposite, are effective in all other component systems of modern
societies except literary systems.
We shall now try to give a short survey of these conventions.
(a) aesthetic convention. It is common knowledge in our society that all
actors in literary systems must be willing and able:
to extend their action potential (or the action potential of other participants
in the literary system) beyond the usual criteria of true/false or useful/useless, and to orient themselves towards expectations, norms, and criteria
which are deemed aesthetically relevant in the respective literary system,
to designate communicative actions intended as literary by appropriate
signals during production, and to follow such signals during reception,
to select as a frame of reference for expressions in literary texts not just the
socially established world model he/she is accustomed to in his/her respective social group but virtually all constructible frames of reference, and
to de-emphasize the fact convention. 6
(b) polyvalence convention. It is common knowledge among all actors in
literary systems in our society that:
-
-
-
-
6 Fact convention. It is c o m m o n knowledge in our society that communicative objects, especially
texts, should permit reference to the world model accepted in that society, such that people can
decide if the assertions conveyed by the text are true and what their practical relevance is.
D. Meutsch, S.J. Schmidt / Understanding literary texts
557
- text producers are not bound by the monovalence convention, 7
- text receivers have the freedom to produce different Kommunikate from the
same Text in different times and situations ( = weak version of the polyvalence convention hypothesis) or in the same reading process ( = s t r o n g
version of the polyvalence convention hypothesisS), and they expect others
to do likewise,
- text receivers rate the realization of different cognitive, emotive, and moral
reading results on different levels of reception as optimal corresponding to
their needs, abilities, intentions, and motivations, though the reasons for
such rating may differ among participants and situations, and
- text-mediators and text-processors should not counteract these regulations
overtly.
The relation between the two conventions might be described as follows:
the aesthetic convention seems to be logically prior to the polyvalence convention. By suspending the true-false and useful-useless frames it enables an
orientation towards poetic norms, expectations, and criteria; it fosters the
multi-readability of literary texts, and enables a subjective optimizing of the
process of Kommunikat-construction: it thus comprises the polyvalence convention.
The polyvalence convention influences the relation between processes of
Kommunikat-construction and the value-system of the reading subject; and
above all it influences the subject's evaluation of the pluri-referentiality of
text-elements and texts (i.e. its poly-readability). Whereas polysemy normally
:nhances the uncertainty of reading processes and is, accordingly, negatively
:onnotated, poly-readability on the basis of the polyvalence convention enlarges the set of reading possibilities and is normally positively connotated.
4. On the development of operational types of cognition in literary understanding
4.1
From the literary-theoretical assumptions outlined in section 3 we derive the
following general hypotheses: due to the aesthetic conoention, problems in the
process of constructing a Kommunikat assigned to a literary text are generally
7 Monovalence convention. It is Common knowledge in our society that: (a) text producers are
expected to shape their texts in such a way that different people at different times can assign them
a constant Kommunikat,and (b) text receiversare expectedto strive for the assignment of a single
Kommunikat to the texts.
s For the differencebetween weak and strong versionsof the polyvalencehypothesissee Groeben
(1982b).
558
D. Meutsch, S.J. Schmidt / Understanding literary texts
not regarded as deficiencies but are taken as triggers for additional 'understanding' efforts. 9
(Such reading problems are at least partially provoked by text features
which - in the reader's implicit or explicit framework of poetical values and
expectations - are deemed literary relevant. However, this point is not topical
in our present project.)
As has been pointed out in section 2, linguists and psychologists convergently assume that understanding processes are governed by strategies. We
hold the view that the polyvalence convention may be regarded as a special
strategy which is relevant in the literary system seen as a component system of
society. Accordingly, everybody who is interested in the analysis of understanding processes can hope to obtain further insight into the 'nature' of
understanding through an analysis of special kinds of unffrrstanding processes,
viz. literary understanding: if our strategy-oriented approach of explaining
text understanding is successful, and if we have to take into account different
types of strategies, then the 'literary case' will be helpful in characterizing the
strategy-type of mediating between the individual and the social domain as the
global sense criterium.
4.2
The hypothetical assumptions connected with the general formulation of the
polyvalence convention can be specified as follows:
(a) Polyvalent processes of Kommunikat-construction (PKC) are triggered by
reading problems which - according to the aesthetic convention - are
positively evaluated.
(b) Polyvalent PKCs can be regarded as a cognitive enlargement of PKC-construction; the enlargement especially concerns the set of frames of reference put to work during the process of Kommunikat-construction.
(c) Polyvalent PKCs imply (presumably meta-cognitive) personal commentaries on reading problems and the construction of enlarged PKCs.
In our empirical work we have tried to examine the three parts of this
hypothesis one after the other, not in a multi-variate procedure.
4.3
From our presupposed theoretical assumption (as presented in sections 2
and 3) and the assumptions connected with them in sections 4.1 and 4.2 we
derive the following hypotheses: our model of understanding works with four
different types of cognition which can be used as operational categories for
deriving special hypotheses, i.e.:
9 See the empirical evidence in Hintzenberg et al. (1980), and Schmidt and Zobel (1980).
D. Meutsch, S.J. Schmidt / Understanding literary texts
559
(1) all descriptions of (details of) the PKC, i.e. the unreflected application of
conventions (abbreviated as KA), lo
(2) all commentaries on the PKC, i.e. the consciously reflected application of
convention (entailing problems of PKC) (abbr. KOM),
(3) all descriptions of relevant conventions (eventually) independent of PKC
(abbr. KB ST), and
(4) all descriptions of back- and forth-operations which are related in terms of
propositions (abbr. VOR and RUE).
These four domains can now be specified as follows:
(1) Descriptions of PKCs
KA PR PO:
Positive propositional instantiations of a thematic slot or
check of thema correspondence.
KA PR NE:
Negative propositional instantiations of a thematic slot or
check of thema correspondence.
KA UR PO:
Positive (non-specific) evaluation of the propositional slot
instantiations resp. thema correspondence.
KA UR NE:
Negative (non-specific) evaluation of the propositional slot
instantiation resp. thema correspondence.
KA RE:
Referring to frames of reference as application of a convention.
KA RE WE:
Changing of frames of reference as application of a convention.
(2) Comments on PKCs
KOM TE PO:
Positive text-concerning comments on PKC.
KOM TE NE:
Negative text-concerning comments on PKC.
KOM RE PO:
Positive reference-frame concerning comments on PKC.
KOM RE NE:
Negative reference-frame concerning comments on PKC.
KOM PE PO:
Positive reader-concerning comments on PKC.
KOM PE NE:
Negative reader-concerning comments on PKC.
(3) The stock of relevant and applicable conventions
LB:
The notion of 'literary'/'literature' specified as to its content.
PRA:
Reading preferences in terms of text types, genres, subjects.
FUN:
Assumptions concerning possible functions of (reading)
literature.
ERW:
Expectations in authors and books.
LESG:
Reading habits.
lo T h e variable n a m e s a n d a b b r e v i a t i o n s were, of course, taken f r o m G e r m a n w o r d s in our
research context.
560
D. Meutsch, S.J. Schmidt / Understanding literary texts
Modes of talking about literary works and readings.
Degree of literary socialization.
GESPR:
LIT SOZ:
(4) Back- and forth-operations
VOR:
Expectations concerning forth-coming text elements.
RUE:
Recorrection of (partial) text readings. 11
5. The hypotheses
In this section we give a detailed report on those hypotheses derived from the
presupposed theoretical framework which we shall put t~ empirical tests. The
criteria for empirical testability are related to the status of concretisation of
the object theory (cf. sections 2 and 3) and to the operational types of
cognition (see section 4). The first hypothesis has already been formulated in
the presupposed empirical theory of literature (Schmidt 1980,1982), it is
well-supported by results of research in literary reading strategies. Carbonell
(1981) assumes that readers interpret understanding problems as authorspecific metaphors which can be understood by 'metaphor-processing'.
11 W e quote e x a m p l e s from the transcribed tape recordings of our test subjects to give the reader
an i m p r e s s i o n of the i n s t a n t i a t i o n of the variables:
Examples
K A PR PO:
NE:
KA U R PO:
NE:
K A RE:
KA RE WE:
KOM TE PO:
NE:
K O M RE PO:
NE:
K O M PE PO:
NE:
VOR:
RUE:
' G e n a u wie die Bl~itter, die von den B~iumen fallen - zwar nicht in den
P a p i e r k o r b k o m m e n , aber in den A b f a l l k o r b . '
' A l s o d a r a u s k a n n ich nicht - das k a n n ich nur so verstehen, d a b gemeint ist,
d a b das, was n o r m a l e r w e i s e auf der Erde stattfindet, nicht e r n s t z u n e h m e n ist,
aber - das k~Snnte m a n j a d u r c h a u s so sehen.'
' D e r Satz gef'~illt mir von den Worten, von den Begriffen her eigentlich sehr
gut.'
' A l s o das hiSrt sich auf e i n m a l sehr b r u t a l an.'
'Ja, d a f'~illt mir schon die Sprache auf ( - ) das erinnert mich so an, an
technische AbRiufe - Haushaltsftihrung.'
' D e n Z u s a m m e n h a n g k~Snnte ich mir auch gut vorstellen - so eine A n a l y s e zur
gesellschaftlichen E n t w i c k l u n g in - aber weg von freiheitlich-demokratischen
Verh~iltnissen - ~ih - zu - sagen wir mal - autorit~iren, d i k t a t o r i s c h e n
Verh~dtnissen.'
' A b e r das ist wahrscheinlich n u r bildlich zu verstehen.'
' A r b e i t s t i s c h - so ein Wort, wo ich auch wieder ein bifAchen dri~ber stolpere in
diesem Z u s a m m e n h a n g . '
' D a s wiirde ich aber auf die N a t u r beziehen.'
'Vielleicht - aber ich verstehe es eben nicht zu dem, was zuletzt war.'
' D a s heist, es f'~illt mir doch was dazu ein.'
' A b e r insgesamt, mul3 ich jetzt sagen, verstehe ich den Satz nicht.'
'Ja, ich hoffe in der n~ichsten Zeile auf B e a n t w o r t u n g dieser Frage.'
' D a s riicke ich auch in die N~ihe von " A b s t i m m u n g s s a a l " , " M a n u s k r i p t e " ,
"Bibliothek".'
D. Meutsch, S.J. Schmidt / Understanding literary texts
561
HI: Reading texts they deem literary, readers usually have understanding
problems.
We operationalize H1 by all the negative dependent variables, i.e. by KA PR
NE, KA UR NE, KOM TE NE, KOM RE NE, and KOM PE NE.
The second hypothesis follows from the presupposed empirical theory of
literature and from the theory (or model) of understanding as outlined in
section 2.
H2: The performance of polyvalent processes of Kommunikat-construction
is realized as a change of frames of reference. ~2
Compared to the 'demetaphorization-strategy' (which can be regarded as the
propositional equivalent of understanding problems) the 'strategy of changing
frames of reference' characterizes the level of schemata (Harker 1981). We
assume that metaphor-processing triggers the superordinated strategy of frames
of reference (cf. Vipond and Hunt 1986), but this is not the subject of our
present project.
H2 is operationalized by KA RE WE.
The third hypothesis, too, has been derived from the presupposed empirical
theory of literature: it is based on the assumption that processes of literary
understanding realize different possible meanings on macrostructural levels
(van Dijk 1980, Meutsch 1986a, Vipond and Hunt 1984).
H3: A change of frames of reference during a process of Kommunikat-construction ~3 which is regarded as an instance of acting according to the
polyvalence convention, is evaluated positively.
We operationalized H3 in terms of (all positive) correlations between KA RE
WE and positive KOMs as well as in terms of the comparison of positive
correlations with negative KOMs.
H4 was derived from the presupposed empirical theory of literature and the
theory of understanding, which in turn is backed by the results of story
grammar research (as concerns the notion of coherence) as well as by the
research in the correlation between thematic (textual) deep structure, coherence,
and understanding (cf. Schmidt and Zobel 1980):
12 In our pilot study we could not check the differences between readings of literary and
non-literary texts. But Meutsch (1986a) has provided evidence for the hypothesis that there
actually exist differences: text subjects try to overcome understanding problems by means of
producing elaborations. (See also Thorndyke 1977, who observed a change of schemata when
confronted by understanding problems.)
13 We have adopted the strong version of the polyvalence convention: see (b) in section 3.2.
562
D. Meutsch, S.J. Schmidt / Understanding literary texts
H4: Non-narrative literary texts (as e.g. poems) trigger more polyvalent
processes of Kommunikat-construction than narrative literary texts (as
e.g. short stories).
In a more implicit way H4 hints at the relation between different kinds of
strategies in literary understanding.
H4 can be checked by location comparison of different texts with regard to
KA RE WE-activities. Secondly, a comparison between a narrative and a
non-narrative text with regard to the sum of all variables attributed to H1 (i.e.
KA PR NE, KA UR NE, KOM TE NE, KOM RE NE, and KOM PE NE)
must favour the non-narrative text. Thirdly, a positive correlation between the
sum of all POS KOM's and KA RE NE's must rank higher for non-narrative
than for narrative texts. The second and third check will only be possible when
the narrativitiy of texts can be clearly defined.
The fifth hypothesis is a derivative of the empirical theory of literature:
H5: The aesthetic convention and the polyvalence convention are interdependent. The aesthetic convention comprises the polyvalence convention. Therefore readers who have accepted the aesthetic convention are
supposed to follow the polyvalence convention as well.
H5 is descriptively operationalized by the KB-categories.
6. T h e empirical test of our h y p o t h e s e s
6.1
In order to test our hypotheses we made use of a two-factorial design (with
text and method as factors). We chose two texts: J. Becker's poem Nab und
fern (which we intuitively regarded as a non-narrative item), and L. Gustafsson's short story Ende des Herbstes (which we intuitively deemed narrative)
(see the Appendix).
We applied the method of thinking aloud while reading sentence by
sentence as well as after having read the whole text. (We thereby tried to
provide a post-hoc check of the still rather vague assumptions concerning the
degree of consciousness of literary cognitions; see Viehoff and Meutsch 1985,
Meutsch and Viehoff 1985). We deliberately decided to work with this
method 14 in order to:
-
preserve as far as possible the flexibility of the process of Kommunikat-construction, i.e. to prevent a narrowing of this flexibility by other methods of
14 This method is explicitly presented in Weidle and Wagner (1982), Definer (1984), Ericsson and
Simon (1979, 1980), OIson, Dully and Mack (1984).
D. Meutsch, S.J. Schmidt / Understanding literary texts
-
563
verbalisation used in psychology, as e.g. recall or recognition (Meutsch
1986a; Meutsch and Viehoff 1985),
provide evaluations of individual and specific reactions (cf. Black, Galambos and Reiser 1984, Meyer and Rice 1982, Olson, Mack and Duffy 1981,
Olson, Duffy and Mack 1984, Schnotz 1984),
take into account the subject-dependency of processes of Kommunikat-construction,
provide a rather ecologically valid and relevant test situation, and
- obtain not only analytical items of comprehension but perhaps also experiental and emotive ones (cf. Spiro 1982).
-
-
Of course we are fully aware of the problem implied in the application of
this method: the verbalisation of processes of Kommunikat-constructions
must not be identified with what 'really' happens in the cognitive domain
during the reading process; test subjects have quite different verbalisation
competences, they are dependent upon motivations as to the degree of
verbalisation etc. (cf. Scheffer 1985, Definer 1984). But nevertheless we hold
the view that at present this method opens up the best way to get information
about what happens in processes of understanding.
Our design enables the test of our five hypotheses and some post-hoc
activities: the categories 1, 2, and 4 are served by a classification of the
stimulated recall protocols, category 3 is served by a questionnaire the test
subjects fill in after the thinking aloud-experiment.
The design we chose for our testing is illustrated in figure 1.
The test was executed from October to November 1985 in Bielefeld, Siegen,
and Mihnster. The subjects (N = 48) were chosen according to the parameters
age, sex, education, and profession. We happened to get a good mixture,
ranging from 15 to 60 years of age, from schoolgirls and housewives to
professors and writers.
We organized a single session for every subject. The subjects received
written instructions which could be further explained if the subjects asked for
it. All sessions were completely recorded on tape. The text was presented
sentence by sentence with free reading time; the subjects were asked to report
on all (cognitive and emotive) processes they became aware of. They were
Method (stimulated recall)
L
Fig. 1.
Commenting sentence by sentence
Commenting after reading
Narrative
text
12 Ss.
12 Ss.
Non-narrative
text
12 Ss.
12 Ss.
564
D. Meutsch, S.J. Schmidt / Understanding literary texts
allowed to go back to previously read sentences (but of course they were
forbidden to skip sentences).
After having finished the comments on (or procotols of) their reading
processes, the subjects were asked to underline those parts of the text which
they deemed literary interesting, beautiful, moving etc. Finally they had to fill
in a questionnaire. The average session took approximately 1.5 hours. D.
Meutsch and S.J. Schmidt acted as test directors; experimenter effects were
not observed. The taped protocols were transcribed sentence by sentence
because we had chosen the sentence as sense unit. Two specially trained raters
assigned the dependent variables to the transcribed sentences. The interrater
correlation was 0.7-0.8. The machine processing was performed by SPSS 9.
We coded all categories mentioned in section 4.3 sentence by sentence in
order to be able to reconstruct both individual reading processes and readings
of different parts of the texts.
7. The results
7.1
It turned out that we got a great variety of data lacking normal distribution.
Accordingly we only made use of non-parametric procedures for checking our
hypotheses concerning correlations and differences in location. Unfortunately,
it turned out, too, that the post-reading thinking aloud groups with both texts
were rather unproductive; accordingly the planned post-hoc-explorations of
the degree of consciousness of literary cognitions could not be realized. The
presupposed theories (or models) did not yet allow for multivariate operations.
The presentation of our results is therefore restricted to descriptions and
non-parametric correlations of 24 subjects. The reported results document the
sum of all variables coded sentence by sentence as counting item for each
subject.
The results with respect to our hypotheses are shown in table 1.
With regard to H2, .~ = 2.7 with range = 9 is important for KA RE WE. We
found a significant ( p < 0.001) positive correlation r s = 0.52 (Spearman) between KA RE WE and the sum of all positive commentaries on processes of
Kommunikat-construction (KOMs). The correlation between KA RE WE and
the sum of all negative commentaries of PKC is somewhat lower (r X = 0.41)
but still highly significant ( p < 0.001). The location comparison of KA RE
WE concerning text 1(= narrative) and text 2 ( = non-narrative) provides a
highly significant ( p < 0.0083) difference (U = 27.5) in favour of the rank sum
of text 2. t5
15 The correction with respect to given ties lies inside the level of significance. Due to the small
number of subjects (N = 24) we looked for exact test values; these, too, are inside the level of
significance.
D. Meutsch, S.J. S c h m i d t /
U n d e r s t a n d i n g literary texts
©
~Z
o
v
~<
r,i
r-
E~
O
r~
©
o
o
e.
"r.
c)
c5
,¢
565
566
D. Meutsch, S.J. Schrnidt / Understanding literary texts
Concerning H5 we could state only descriptively the existence of conventions as prediced by the empirical theory of literature. Due to the low number
of subjects we could not check the degree or kind of interdependence of the
cognitive operations caused by the aesthetic and the polyvalence convention.
The 13 items in our questionnaire defined to indicate a subject's competence
of following the aesthetic-convention could be found in at least 85% of all
case. (This result equals those of Hintzenberg et al. 1980: 62-70.)
8. Discussion of the results
8.1
Ad HI: we actually found reading problems in connection with all the
dimensions indicated by our dependent variables; that means that H1 has
been confirmed (in relation to the restricted conditions of our pilot study).
We found an interesting relation between PKC-commentaries (KOMs) and
uncommented PKCs (KAs). A comparison of negative and positive KAs yields
a dominance of positive KAs whereas a comparison of the two kinds of KOMs
shows a dominance of negative KOMs (see table 1). An exact interpretation of
this result presupposes a clarification of the individual function of KAs and
KOMs in relation to the whole PKC; otherwise we can only speculate. With
regard to understanding problems of which the subjects appear to be fully
aware, references to positively experienced processes prevail. But this results
cannot be generalized for all literary texts; in order to do that we have to
check quite divergent literary texts and contrast them with non-literary texts
(cf. Schmidt 1985). In addition, we do not know how far understanding
problems can be regarded as the reason or motive of literary understanding
processes. In case this assumption holds true, literary understanding processes
would have to be explained in terms of a special kind of disambiguation (cf.
Schmidt 1985, Viehoff and Schmidt 1985; Viehoff and Meutsch 1985).
8.2
Ad H2: in the groups of subjects who commented on their reading processes
sentence by sentence, we ascertained a mean of 2.7 changes of frames of
reference per reader and text. If this result can be confirmed in future tests this
would mean considerable empirical support of the strong version of the
polyvalence hypothesis (sensu Schmidt 1980). We did not interpret the relation
between positive and negative comments, because - due to the small random
test - we executed no multivariate analysis. This restriction continues our
remarks in section 8.1: our present project does not provide evidence for
details of the process of literary understanding, its components, and the
D. Meutsch, S.J. Schmidt / Understanding literary texts
567
relations between the components. We could only grasp certain isolated
components. In addition, we need a control group (working with non-literary
texts) in order to provide proper findings concerning changes of frames of
reference (cf. experiment 2 in Meutsch 1986a, see also Meutsch 1986b).
8.3
Ad H3: the results show that a change of frames of reference triggered by the
text is indeed positively evaluated. The correlation between KA RE WE and
all negative KOMs is lower than the one between KA RE WE and all positive
KOMS, but it is still highly significant. This finding is 'heuristically' productive; but is relativizes the result of H3 which is not unequivocally confirmed.
Our analysis of the correlation is still too global to allow for causal explanations. In a further test we plan to combine an analysis sentence by sentence
with an analysis of whether or not changes of frames of reference are deemed
to be of literary relevance.
8.4
Ad H4: non-narrative literary texts actually seem to trigger more polyvalent
processes of Kommunikat-construction than narrative texts. The application
of the U-test yields a highly significant dominance of the poem with respect to
changes of frames of reference. Further projects will have to provide more
evidence for the assumption that the changing of frames of reference is a
necessary condition for understanding processes of literary texts. According to
available empirical results (see Laszlo 1985) the existence of narrative structures in literary texts cannot be seen as a diminution of their poeticity.
The additional checks for H4 (see section 5) did not yield significant results.
As a consequence we intend to differentiate the factor 'text' on the basis of
explicit theories in future projects.
8.5
Ad H5: this hypothesis has been fully confirmed (,~75%). This result corresponds to the one Hintzenberg et al. (1980) found in a survey of the present
population in the FRG.
Due to the structure of the questionnaire and the low number of subjects
we could not analyse the stock of aesthetic-convention indicators subject by
subject. Further projects have to deal with the following problems:
(a) The procedures used by Hintzenberg et al. and by ourselves to describe the
form of the aesthetic-convention, are valid and reliable with regard to their
purely descriptive function. Interpretations intending to make use of this
568
D. Meutsch, S.J. Schmidt / Understanding literary texts
stock as a covariate have to develop new modes of constituting this
convention (cf. Schmidt 1985).
(b) As concerns possible functions of literary reception (guided by the aesthetic
convention) the results of applying the questionnaire should not be generalized. An improvement presupposes a clarification of problems as discussed in the next paragraph.
9. G e n e r a l d i s c u s s i o n
(1) The description of PKC needs further specification. We have to regard
smaller units than sentences in order to be able to account for interactions
between micro- and macropropositional processes. On the other hand, specification means a more precise explication of qualitative levels of action in the
understanding process. The most popular terms for characterizing literary
understanding ('polysemy') and non-literary understanding ('coherence') do
simply not account for these levels. As long as we are not able to provide a
theoretical model which explicitly displays different types of sense constancy
that are (can be) used in understanding, we still fail to clarify text reception.
Any exact analysis of interactive process (of different ranks) in verbal understanding presupposes a clarification of the relation between models of both
language description and action description. Otherwise we shall run into
psychologically naive text ontologies or into linguistically naive psycholinguism.
(2) With regard to the models of data evaluation 16 we have to look for
explicit models of text understanding of a holistic format (cf. Schnotz 1985,
Johnson-Laird 1983, Collins, Brown and Larkin 1980) which allows us to
apply multivariate procedures. Until such a model is available we remain
restricted to procedures of the kind we have made use of in the present
project.
(3) As concerns the further methodological development two approaches
seem to be promising (Meutsch and Viehoff 1986):
- O n e group of researchers favours consensus-theoretical approaches where
the test subject acts as correcting instance. In this approach both the
complexity and the globality is increased - with all pros and cons connected with such a decision (cf. e.g. the procedure of Scheele and Groeben
1984).
- Another group tries to increase the complexity of their theories, making use
of multivariate models. They, too, tend to relativize strict monocausal
explanations, but then still advocate the use of standardized psychological
test methods (cf. e.g. Meutsch 1986a).
16 F o r a critical analysis of the handling of our project see Alfes (1985).
D. Meutsch, S.J. Schmidt / Understanding literary texts
569
In our view this alternative should not be i n t e r p r e t e d in terms of an
e i t h e r - o r of h a r d vs. soft m e t h o d s , b u t i n s t e a d as a p l e a for c o o p e r a t i o n a n d
c o m p l e m e n t a r i t y in the field of m e t h o d o l o g y .
I0. Appendix
Gedicht (Autor: Jiirgen Becker)
N A H U N D FERN
in Gruppen yon Blacken
Erscheinendes spricht so wie bezeichnet oder zu nennen
ansteigend das tiefliegende Panorama
Reihung am Horizont
und das Ausbleiben einer Empfindung heute abend
ein Kamm aug der Tasche gefallen
wie so oft wie verf'tigbar ich war hemmungslos
dringeblieben im Betrieb
rote Streifen sagen wie GewiSlk
sp~iter eine Schleife dunkler Friihlingsvtigel
die Schwellungen dazu des braunen Bodens
unsichtbar die Buchstaben der Seufzer
weitere beleuchtete Gruppen
Bewegungen in der Ferne vielleicht neue Menschlichkeit
jedenfalls die Ebene neuer Gebetsinhalte
jedenfalls in der Erwartung des Cups
oder was Interesse
oder gilt nichts im Ger~usch der Menge
sagen wit Formen der Hoffnung
leuchtende IdentitSt
und nicht wie unten das Zucken der Ampeln
570
D. Meut'sch, S.J. Schmidt / Understanding literary texts
Prosa (Autor: Lars Gustafsson)
DAS ENDE DER HERBSTES
Der ganze Herbst ist zum SchluB nut noch ein kalter Aufgul3. Das tote Laub aller
Baumarten wird im Regen eingeweicht. Keine Garung, kein Entstehen von Alkohol:
erst im Fri~hjahr wird es sich zeigen, ob die kalten Umschl~ge an einem Holzbein
wirken.
Das Entbl~ittern geschieht liederlich. Alle Ti~ren des Abstimmungssaales springen
auf, schlagen zu. In den Papierkorb, in den Papierkorb! Die Natur zerreil3t ihre
Manuskripte, verwi~stet ihre Bibliothek, schl~gt im J~ihzorn ihre letzten F ~ c h t e mit
einer Stange ab.
Dann erhebt sie sich unvermittelt von ihrem Arbeitstisch. Gleich wirkt ihre Statur
riesenhaft. Ungek~immt tr~igt sie das Haupt im Nebel. Mit herabh~ingenden Armen
zieht sie geniel3erisch den eisigen Wind ein; der mach ihr den Kopf wieder klar. Die
Tage sind kurz, die Nachte brechen schnell herein, die Heiterkeit hat ausgespielt.
Wie die anderen Gestirne in der Luft wird die Erde nun wieder ernsthaft. Ihre helle
Seite ist schmaler, durchfurcht von Schattenl~itern. Ihr Schuhwerk saugt sich voll
Wasser und quietscht wie bei einem Vagabunden.
In diesem Froschgesumpfe, dieser heilsamen amphibischen Zweideutigkeit, kommt
alles wieder zu Kr~iften, springt von Stein zu Stein, wechselt von einem Feld zum
anderen. Der Rinnsteine werden mehr und mehr.
Ein gelungenes Grol~reinemachen, und ohne Ri~cksicht auf Konventionen! Ob
angezogen oder nackt - bis auf die Knochen durchn~il3t.
Und das dauert seine Zeit, trocknet nicht etwa sofort. Drei Monate heilsamen
Nachdenkens in diesem Zustand; ohne dab die Gef~il3e drauf reagieren, ohne Bademantel, ohne Rol3haarhandschuhe. Doch ihre kraftige Konstitution i~bersteht alles.
Wenn nun die kleinen spitzen Knospen wieder hervorkommen, so wissen sie, was sie
tun und warum es wieder losgeht, - und wenn sie sich nur vorsichtig herauswagen, vor
K~ilte starr und mit rotem Gesicht, so kennen sie den Grund dafter.
Aber hier beginnt eine neur Geschichte, die vielleicht daraus folgt, aber nicht so
riecht wie das schwarze Lineal, mit dem ich jetzt einen Schlul3strich unter diese hier
ziehen will.
D. Meutsch, S.J. Schmidt / Understanding literary texts
571
References
Alfes, H., 1985. 'Lautes Denken' als Methode im Rahmen des Polyvalenz-Pilotprojekts:
l~lberlegungen und kritische Anmerkungen zu Methodenanwendungen und Protokollauswertungen. (Unpublished working paper, University Siegen.)
Anderson, J.R., 1985. Cognitive psychology. New York: Freeman.
Anochin, P.K., 1967. Das funktionelle System als Grundlage der physiologischen Architektur des
Verhaltensaktes. Jena: Fischer.
Ballstaedt, St.-P., H. Mandl, W. Schnotz und S.O. Tergan, 1981. Texte verstehen. Texte gestalten.
Miinchen, Wien, Baltimore: Urban & Schwarzenberg.
Bates, E. and B. MacWhinney, 1979. 'A functionalist approach to the acquisition of grammar'. In:
E. Ochs and B. Schieffelin, eds., 1981. Developmental pragmatics. New York: Academic Press.
pp. 167-211.
Black, J.B., J.A. Galambos and B.J. Reiser, 1984. 'Coordinating discovery and verification
research'. In: D.E. Kieras and M.A. Just, eds., 1984. New methods in reading comprehension
research. Hillsdale, NJ: Erlbaum. pp. 287-297.
Carbonnell, J.G. Jr., 1981. Metapher comprehension. Technical report, Department of Computer
Science, Carnegie-Mellon University.
Clark, H.H., 1977. 'Inferences in comprehension'. In: D. La Berge und S.J. Samuels, eds., 1977.
Basic processes in reading: perception and comprehension. Hillsdale, N J: Erlbaum. pp.
243-263.
Collins, A., J.S. Brown and K.M. Larkin, 1980. 'Inference in text understanding'. In: R.J. Spiro,
B.C. Bruce and W.F. Brewer, eds., Theoretical issues in reading comprehension. Hillsdale, N J:
Erlbaum. pp. 385-407.
Cranach, M. von, U. Kalbermatten, K. Indermiihle und B. Gugler, 1980. Zielgerichtetes Handeln.
Bern, Stuttgart, Wien: Huber.
Cranach, M. von, 1983. 'Uber die bewuBte Repr~isentation handlungsbezogener Kognitionen'. In:
L. Montada, K. Reussere und G. Steiner, Hrsgs., Kognition und Handeln (Festschrift H.
Aebli). Stuttgart: Klett-Cotta. 64-92.
Crothers, E.J., 1979. Paragraphs, structure, inference. Norwood, NJ: Ablex.
Definer, G., 1984. Lautes Denken - Untersuchung zur Qualitat eines Datenerhebungsverfahrens.
Frankfurt, Berlin, New York: Lang.
Dijk, T.A. van, 1979. Cognitive processes of literary discourse. Poetics Today 1, 143-159.
Dijk, T.A. van, 1985. Handbook of discourse analysis. New York: Academic Press.
Dijk, T.A. van and W. Kintsch, 1983. Strategies of discourse comprehension. New York:
Academic Press.
D~Srner, D., 1976, Probleml/Ssen also Informationsverarbeitung. Stuttgart: Kohlhammer.
D~Srner, D., 1983. 'Kognitive Prozesse und die Organisation des Handelns'. In: W. Hacker, W.
Volpert und M. v. Cranach, Hrsg., 1983. Kognitive und motivationale Aspekte der Handlung.
Bern, Stuttgart, Wien: Huber. pp. 26-37.
Ericsson, K.A. and H.A. Simon, 1979. Thinking-aloud protocols as data: effects of verbalization.
Carnegie-Mellon University, Dept. of Psychology (CIP working paper no. 397).
Ericsson, K.A. and H.A. Simon, 1980. Verbal reports as data. Psychological Review 87, 215-251.
Flammer, A. and W. Kintsch, eds., 1982. Discourse processing. Amsterdam: North-Holland.
Glasersfeld, E. von, 1986. Begriffssemantik und Wissenskonstruktion. Braunschweig, Wiesbaden:
Vieweg. (In print.)
Graesser, A.C., 1981. Prose comprehension beyond the word. New York: Springer.
Graesser, A.C., S.E. Gordon and J.D. Sawyer, 1979. Memory for typical and atypical actions in
scripted activities: test of a script pointer+ tag hypothesis. In: Journal of verbal learning and
verbal behaviour 18, 319-332.
Groeben, N., 1982a. Leserpsychologie: Textverstiindnis - Textversffmdlichkeit. Miinster:
Aschendorff.
572
D. Meut'sch, S.J. Schmidt / Understanding literary texts
Groeben, N., 1982b. Methodologischer AufriB der empirischen Literaturwissenschaft. SPIEL 1,
26-89.
Hacker, W., 1978. Allgemeine Arbeits- und Ingenieurpsychologie. Bern, Stuttgart, Wien: Huber.
Hacker, W., Hrsg., 1980, Optimierung kognitiver Arbeitsanforderung. Bern, Stuttgart, Wien:
Huber.
Harker, W.J., 1981. Comprehending the discourse of poetry. Beitrag zum International Symposium on Text Processing at Fribourg, Switzerland, September 1981.
Hauptmeier, H. und S.J. Schmidt, 1985. Einftihrung in die empirische Literaturwissenschaft.
Braunschweig, Wiesbaden: Vieweg.
Hintzenberg, D., S.J. Schmidt und R. Zobel, 1980. Zum Literaturbegriff in der Bundesrepublik
Deutschland. Braunschweig, Wiesbaden: Vieweg.
HiSrmann, H., 1976. Meinen und Verstehen. Grundz0ge einer psychologischen Semantik. Frankfurt/M: Suhrkamp.
H/Srmann, H., 1983. 'On the difficulties of using the concept of a dictionary - and the
impossibility of not using it'. In: G. Rickheit and M. Bock, eds., Psycholinguistic studies in
language processing. Berlin, New York: de Gruyter. pp. 3-16.
Hoppe-Graff, S., 1984. 'Verstehen als kognitiver ProzeB. Psychologische Ans~itze und Beitr~ige
zum Textverstehen'. In: W. Klein, Hrsg., Textverst~indlichkeit - Textverstehen. G/3ttingen,
Vandenhoeck & Ruprecht. pp. 10-37. ( = Lili. Zeitschrift ftir Literaturwissenschaft und
Linguistik 55, 1984.)
Johnson-Laird, P.N., 1983. Mental models. Towards a cognitive science of language, inference and
consciousness. Cambridge: Cambridge University Press.
Just, M.A. and P.A. Carpenter, 1980. A theory of reading: from eye fixations to comprehension.
Psychological review 87, 329-354.
Kintsch, W., 1978. 'Comprehension and memory of text'. In: W.K. Estes, ed., Handbook of
learning and cognitive processes, Vol. 6. Hillsdale, N J: Erlbaum.
Kintsch, W. and E. Greene, 1978. The role of culture specific schemata in comprehension and
recall of stories. Discourse Processes 1, 1-13.
K/ix, F., 1980. Erwachendes Denken. Berlin: VEB Verlag der Wissenschaften.
Kluwe, R., 1979. Metakognition. Miinchen: Psychologisches Institut der Universit~it.
Laszlo, J., 1985. Different points of view in story processing. Paper for the IX International
Colloquium on Empirical Aesthetics, Santa Cruz (CA), USA, August 1985.
Leont'ev, A.N., 1979. T~itigkeit, BewuBtsein, Pers/Snlichkeit. Berlin: Aufbau.
Mandl, H. and W. Schnotz, 1985. New directions in text processing. Forschungsbericht 36 des
Deutschen Instituts for Fernstudien an der Universit~it Ttibingen. TObingen: DIFF.
Mandl, H., N. Stein and T. Trabasso, eds., 1984. Learning and comprehension of text, Hillsdale,
N J: Erlbaum.
Mandler, J.M. and N.S. Johnson, 1980. On throwing out the baby with the bathwater: a reply to
Black & Wilensky's evaluation of story grammars. Cognitive Science 4, 305-312.
Maturana, H.R., 1982. Erkennen: Die Organisation und Verk~Srperung von Wirklichkeit.
Braunschweig, Wiesbaden: Vieweg. (Wissenschaftstheorie. Wissenschaft und Philosophie, Bd.
19.)
McCormick-Leighty, K. 1985. Psychological realism: a new epistemology for reader-response
criticism. Paper for the Carnegie-Mellon-Conference in April 1985.
Meutsch, D., 1984. 'Wie "entsteht" ein verst~indlicher Text? Einfl0sse literarischer und nichtliterarischer Kontexte auf zielspezifische Verstehensprozesse'. In: W. Klein, Hrsg., 1984.
Textverst~tndlichkeit - Textverstehen. G~Sttingen: Vandenhoeck & Ruprecht. pp. 86-112.
( = Lilie, Zeitschrift for Literaturwissenschaft und Linguistik 55/84.)
Meutsch, D., 1986a. Literatur verstehen. Eine empirische Studie. Braunschweig, Wiesbaden:
Vieweg. (In print.)
vleutsch, D., 1986b. Speculations about a model of literary discourse processing: towards the
integration of textual and psychological explanation. Poetics 15. (Forthcoming.)
D. Meutsch, S.J. Schmidt / Understanding literary texts
573
Meutsch, D. and R. Viehoff, 1985. Types of inference and elaboration in literary understanding.
Paper for the IX International Colloquium on Empirical Aesthetics, Santa Cruz (CA)~ USA,
August 1985.
Meutsch, D. and R. Viehoff, eds., 1986. Models of literary understanding. Berlin-New York: de
Gruyter. (In print.)
Meyer, B,J.F., 1984. 'Text dimensions and cognitive processing', In: H. Mandl, N.L. Stein and T.
Trabasso, eds., 1984. Learning and comprehension of text. Hillsdale, NJ: Erlbaum.
Meyer, B.J.F. and G.E. Rice, 1982. The interaction of reader strategies and the organisation of
texts. TEXT 2, 155-192.
Miller, G.A., E. Galanter and K.H. Pribram, 1960. Plans and the structure of behaviour. London:
Holt. Rinehart & Winston.
Neisser, U., 1976. Cognition and reality. San Francisco: Freeman.
Olson, G.M., S.A. Duffy and R.L. Mack, 1984. 'Thinking-out-loud as a method of studying real
time comprehension processes'. In: D.E. Kieras and H.A. Just, eds., New methods in reading
comprehension research. Hillsdale, NJ: Erlbaum. pp. 253-286.
OIson, G.M., R.L. Mack and S.A. Duffy, 1981. Cognitive aspects of genre. Poetics 10, 283-315.
Rickheit, G. and H. Strohner, eds., 1985. Inferences in text comprehension. Amsterdam: NorthHolland.
Rumelhart, D.E., 1980. 'Schemata: the building blocks of cognition'. In: R. Spiro, B. Bruce and
W. Brewer~ eds., 1980. Theoretical issues in reading comprehension. Hillsdale, N J: Erlbaum.
Schank, R.C. and M. Lebowitz, 1980. Levels of understanding in computers and people. Poetics 9,
251-273.
Scheele, B. and N. Groeben, 1984. Die Heidelberger Struktur-Lege-Technik (SLT), Eine DialogKonsens-Methode zur Erhebung subjektiver Theorien Mittlerer Reichweite, Weinheim &
Basel: Beltz.
Scheffer, B., 1985. Literatur als Selbstkommunikation. Universit~it Bielefeld. (Habilitationsschrift.)
Schmidt, S.J., 1980. GrundriB der Empirischen Literaturwissenschaft. Teilband 1: Der Gesellschaftliche Handlungsbereich Literatur. Braunschweig, Wiesbaden: Vieweg. (Konzeption Empirische Literaturwissenschaft, Bd. 1.)
Schmidt, S.J., 1982. Foundations for the empirical study of literature. (Translated and fully
revised by R. de Beaugrande.) Hamburg: Buske.
Schmidt, S.J., 1983. 'Text, Subjekt und Gesellschaft. Aspekte einer konstruktivistischen Semantik.'
In: M. Faust et al., Hrsg., Allgemeine Sprachwissenschaft, Sprachtypologie und Textlinguistik.
Festschrift ft~r P. Hartmann, T'Obingen: Narr. pp. 55-71.
Schmidt, S.J., 1984. Empirische Literaturwissenschaft in der Kritik. SPIEL 3, 291-332.
Schmidt, S.J., 1985. Texte verstehen - Texte interpretieren. (In print.)
Schmidt, S.J. und R. Zobel, 1980. Textkomplexit~it und Leserverhalten. Braunschweig, Wiesbaden:
Vieweg.
Schnotz, W., 1984. Textaufbau und Koh~irenzbildung. Forschungsbericht des DIFF Nr. 24,
Ti~bingen, Februar 1984.
Schnotz, W., 1985. Elementaristische und holistische Theorieans~tze zum Textverstehen. Forschungsbericht 35 des Deutschen Instituts f'tir Fernstudien an der Universit~it Tiabingen.
Tiabingen: DIFF.
Spiro, R.J., 1982. Long-term comprehension: schema-based vs. experimental and evaluative
understanding. Poetics 11, 72-86.
Stadler, M., P. Schwab und Th. Wehner, 1979. 'Kognition als Abbild und Plan des Handelns'. In:
H. Ueckert und D. Rhenius. Hrsgs., Komplexe menschliche Informationsverarbeitung. Bern,
Stuttgart, Wien: Huber. pp. 38-46.
Tergan, S.-O.. 1984. Diagnose yon Wissenstrukturen. Forschungsbericht Nr. 30 des Deutschen
Instituts for Fernstudien an der Universitat Ttibingen. T't~bingen: DIFF.
Thorndyke, P.W., 1977. Cognitive structures in comprehension and memory of a narrative
discourse. Cognitive Psychology 9, 77-110.
574
D. Meutsch, S.J. Schmidt / Understanding literary texts
Verdaasdonk, H., 1982. Conceptions of literature as frames? Poetics 11, 87-104.
Viehoff, R., 1976. 'Uber einen Versuch den Erwartungshorizont zeitgenrssischer Literaturkritik
empirisch zu objektivieren'. In: N. Groeben, Hrsg., Literaturpsychologie. Grttingen:
Vandenhoek and Ruprecht. pp. 96-124.
Viehoff, R., 1982. Do children really understand texts as literary texts? in: Poetics 11, 345-369.
Viehoff, R., 1983. Rezeption und Verarbeitung. SPIEL 2, 101-121.
Viehoff, R. and D. Meutsch, 1985. The method of thinking aloud in the investigation of literary
discourse. Paper for the IX International Colloquium on Empirical Aesthetics, Santa Cruz
(CA), USA, August 1985.
Viehoff, R. and S.J. Schmidt, 1985. Kommunikatbildungsprozel3. Empirische Untersuchungen zur
Struktur und Function konventionsorientierten literarischen Wissens. Antrag an die DFG ft~r
das Schwerpunktprogramm 'Wissenspsychologie'. Februar 1985. (Manuscript.)
Vipond, D. and R.A. Hunt, 1984. Point-driven understanding: pragmatic and cognitive dimensions of literary reading. Poetics 13, 261-277.
Vipond, D. and R.A. Hunt, 1986. 'Dimensions of literary reading'~In: D. Meutsch and R.
Viehoff, eds., Models of literary understanding. New York: de Gruyter. (In print.)
Volpert, W., 1983. 'Psychologische Handlungstheorie - Anmerkungen zu Stand und Perspektive'.
In: W. Volpert, Hrsg., Beitr~ige zur psychologischen Handlungstheorie. Bern, Stuttgart, Wien:
Huber. pp. 13-27.
Warren, W.H., D.W. Nicholas and T. Trabasso, 1979. 'Event chance and inferences in understanding narratives'. In: R.O. Freedle, ed., New directions in discourse processing. Norwood.
NJ: Ablex. pp. 22-52.
Weidle, R. and A.C. Wagner, 1982. 'Die Methode des lauten Denkens'. In: G.L. Huber and H.
Mandl, Hrsg., Verbale Daten. Weinheim: Beltz. pp. 81-103.