INTEGRATING QUALITATIVE REASONING AND TEXT PLANNING 
TO GENERATE CAUSAL EXPLANATIONS 
Farm CERBAH 
DEA/IA 
Dassault Aviation 
78, quai Marcel Dassault 92214 St-Cloud FRANCE 
Tel: (33-1) 47 11 53 {30; Fax: (33-1) 47 11 52 83 
Universit6 de Caen 
I - INTRODUCTION 
Several works IMcKeowu 86, Suthers 88\] have empha- 
sized the common aspects of Explanation Production in ex- 
pert systems and Text Generation. The work described in 
this paper deals with text generation applied to a particular 
type of explanations: causal explanations of physical sys- 
tems. They have akeady motivated influential develop- 
ments in the field of qualitative reasoning about physical 
systems. A central goal of the theories developed in \[De 
Kleer 84\] and \[Forbus 84\] was to elaborate conceptual 
frameworks for providing causal accounts of physical sys- 
tems, sensitive to our commonsense understanding of the 
physical world. Those qualitative causal models constitute 
an adequate starting point as we are interested in how peo- 
ple present such causal explanations in verbal form. 
We will describe our approach for text generation, based ou 
the study of texts collected in encyclopedia and textbooks, 
and currently developed in a system intended to be asso- 
ciated to the qualitative simulation system SQUALE 
\[J6z6quel & Zimmer 92\]. Our conceptual model, which 
constitutes the input to the text generation process, is based 
on Qualitative Process Theory IForbus 84\]. 
According to the "traditional" division of tasks in text 
generation, the transition from conceptual representation 
of causal behaviour to causal explanation in natural lau- 
gouge is viewed as a three-stage process: content specifica - 
tion, text organization and surface generation. 
The content specification task aims at posting communica- 
tive goals described by means of communicative acts on 
conceptual entities. In particular, the causal explanation to 
be produced is often restricted to ,some particular events of 
the causal behaviour. We will show how relevant informa- 
tion and appropriate communicative acts are identified. 
Text organization is the most elaborate part of our model 
and is also divided into three tasks. 
The first is concerned with the construction of a textual 
structure from a set of communicative acts established dur- 
ing content specification. This structure, which takes an in- 
termediary place between communicative acts and surface 
realizations, specifies essentially prescriptions on group- 
ing and ordering of textual units. This process is achieved 
through the application of discourse strategies which con- 
trol local Iransitions from communicative acts to possible 
organizational preseriptions. We dcseribe three strategies 
used for structuring causal explanations: a causal chain 
strategy (for organizing simple causal chains), a parallel 
strategy (to impose a parallel structure on the text), and a 
concessive strategy (for performing concessive acts). The 
second task segments the textual structure into sentential 
contents. Several factors are revolved, mainly communica- 
tive form of textual relations and number of textual unite 
expressed in a sentence. 7'he stylistic adjustment task deals 
with stylistic considerations related to lexico-syntactic 
choice, s. We want to ensure an inter-sentential compatibili- 
ty, from a stylistic point of view, between linguistic realiza- 
tions. 
Concerning surface generation, two types of systems have 
been considered: a sentence geuerator \[Fournier 91\] based 
on the Meaning-Text Theory \[Mel'cuk 88a\] and a syntactic 
component based on equivalenee classes between sentence 
structures. 
Our approach suggests a clear separation of knowledge 
sources involved in each of these stages. We will start with 
a presentation of the conceptual model, where we propose 
a characterization of causal interactions betweeo physical 
events (essentially, physical parameter changes). The next 
sections deal with the different stages of the text generation 
process. 
AcrEs DE COLING-92, NA~res, 23-28 no(n' 1992 6 1 7 PRoc. OF COLING-92, NANTES, AUO. 23-28, 1992 
H- THE CONCEPTUAL FRAMEWORK 
Qualitative Process Theory helps to capture commonsense 
knowledge implicit in mathematical models of traditional 
physics. Besides, it provides an ontological framework for 
expressing physical causality, by expliciting the physical 
mechanisms. In order to describe physical situations, two 
kinds of structures are used : individuals views, for repre- 
senting physical objects and behavioural abstractions, and 
physicalprocesses such as fluid and heat flows. Gradual 
parameters are associated to identified entities (individual 
views and processes) and causal dependencies are ex- 
pressed by means of two types of qualitative relations: in- 
fluences, for direct effects of processes, and qualitative 
proportionalities, which propagate direct effects of pro- 
eesses on the other parameters. It follows that the initial 
cause of a parameter change is always a process. 
Qualitative relations can aim at opposite effects. In die 
physical situation of figure 1, the mercury expansion tends 
to increase the mercury level while the tube expansion 
tends to lower it. The first influence is predominant, since 
an increase of the mercury level is observed in that situa- 
tion, and thus the second is inhibited 1. In order to represent 
different types of causal interactions, we have introduced 
the causal primitives cause, contrary-cause and opposite- 
effects which are defined according to influences between 
events: 
,, cause: causal relation between two events linked with a 
predominant influence (for example, mercury expansion 
is the cause of mercury level rise) 
• contrary-cause: contrastive relation between two 
events linked with an inhibited influence. One of the 
events is the "contrary cause" of the other (tube expansion 
is a contrary cause of mercury level rise). 
• opposite-effects: contrastive relation between two 
events which aim at opposite effects (mercury expansion 
and tube expansion). 
The last two primitives are called causal opposition rela- 
tions. In that framework, physical system behaviours are 
represented with influence graphs, where physical events 
are linked with causal primitives. An influence graph en- 
closes all events and causal interactions identified in the 
given physical situation. 
I11 - CONTENT SPECIFICATION 
The content specification is primarily concerned with se- 
lection of relevant physical events and causal interactions 
from the influence graph. But the problem should not be 
narrowed to factual knowledge selection, for the conceptu- 
1. It does not mean that the mercury expansion has no effect 
on the mercury level but that no effect is perceived at that 
level of a~tr~ction. 
ul content of the text cannot be considered independently 
of the speaker's communicative intentions. Communica- 
tive roles assigned to conceptual entities have crucial im- 
plications at each stage of the text generation process. 
Hence, they should be made explicit at the content specifi- 
cation level as well as their related conceptual units. In our 
model, the content specification task results in a setofcom- 
municafive acts, which represents the illocutionary struc- 
ture of the message. These acts are performed so as to pro- 
duce some effects on the hearer, as modifying his belief 
system or affecting his emotional states. They may be char- 
acterized according to these in tended effects. We foc us here 
on three types of acts: inform, explain and concede, and es- 
pecially in the context of causal explanation production. 
Roughly speaking, an act of informing aims at making the 
hearer believe some proposition while an act of explaining, 
viewed as a discourse act, has the effect of Supporting a pre- 
viously introduced proposition in order to favour its accep- 
tance. (1) and (2) are respective realizations of acts of in- 
forming and explaining: 
(1) La dilatation du mercure provoque la hausse du 
niveau de mercure, 
(The mercury expansion causes the rise of the mercury 
level.) 
(2) Le niveau de mercure monte, paree que le mercure se 
dilate. 
(The mercury level rises, because of the mercury 
expansion.) 
(1) is a possible achievement of the act: 
inform( cause( D s l volume( mercury ) \] = l , 
Dsllevel(mercury)\]=l )) 2 
and (2), of both acts: 
inform(Dsllevel(mercury)\] = 1 )) 
explain(Ds\[level(mercury)\] =1 ,Ds\[volume(mercury)\] =1) 
In terms of speech acts, (1) i~ a single assertion, which in- 
forms the hearer of the causal connection between the two 
events. In contrast, (2) contains two assertions, the second 
assuming an explicative function. Note that in beth eases 
the same conceptual content is involved. Some differences 
between sentences (1) and (2) as the order of information 
units are due to distinct communicative role assignments. 
We will show in the next .section how these intended "rhe- 
torical" effects influence choices at the text organization 
level. 
But now let us turn to the content specification procedures. 
The determination of communicative acts highly depends 
on the problem solving activity (here, qualitative simula- 
2. In the formalism of Qualitative Process Theory, the Ds 
function provides the derivative sign of the parameter given 
as argument. Ds\[X\]=-I means that X decreases, Ds\[X\]=0 
that X is steady and Ds\[X\]=I that X increases. 
Ac'rns DE COtiNG-92, NANTES, 23-28 AOt\]'r 1992 6 1 8 PROC. OF COLING-92, NANTES, AUG. 23-28, 1992 
cause ~rary--cause cause w 
contained-liquid(m, tube, mercury) Ds\[temperature(w)\]=-I Ds\[tempe\[ature(tube)\]=l Ds\[tempe\[ature(m)\]=l 
liquid(w, water) 
heatjlow(hfl, w, tube) ¢~-/°Pp°slte-~n'ea~Ts~ 
heat-flow( hf2 , tube, m) D s\[ volume( tube )\] =1 D s lvolume(m) \]= l 
talnlrary~cause cause 
fig I : An example of physical system (a thermometer immersed in hot water ) and 
its behavioural description with an influence graph 
inform(hf2) 
inform( cause(hf2,Ds\[ volume( m ) \]= l ) ) 
explain(Ds\[volume(m)\]= 1 ,Ds\[temperature(m I)= 1 ) 
inform(cause(Ds\[volume(m)\]= l,Ds\[level(m)\]= 1 )) 
~==I====D 
Ds \[temper= 1 hf2 J 
cause-effect~se-e~ect(equi) 
Ds\[volume(m)\]= 1 I 
cause-effect(equi) 
Ds\[le!el(m)\]=l 
B y a transfert de chaleur du tube vers le mercure. En consdquence, le mercure se dilate, car sa temperature 
augmente. Cette dilatation provoque la hausse du niveau tie mercure. 
(There is heat flow from the tube to the mercury. As a consequence, the mercury expands,for its temperature 
rises. This expansion causes the mercury level to rise.) 
fig 2 : Application of the causal chain strategy 
hfl I 
cause~effect(equi) 
inform(cause(hfl ,Ds\[temperature(w)\]~ I )) 
inform(cause(hfl,Ds\[temperature(tube)\]=l)) ~r a~=l~ explain(Ds\[tempemture(w)\]=-l,Ds\[heat(w)\]~l) ~ ature(w)l=-I Ds\[ternper (tube\]) 
exp/ain(Ds\[temperature(tube)\]= 1,Ds\[heat(tube)\]= 1) 
cause-effect(pre)/ cause-effe~pre) 
Ds\[h~t(w)\]=-I Ds\[heat(Tube)\]=l 
II y a transfert de chaleur de l'eau vers le tube. En consequence, la temperature de l'eau diminue, 
car sa quantitE de chaleur diminue, et celle du tube augmente, car sa quantitE de chaleur augmente. 
(There is a heat flow from the water to the tube. As a consequence, the temperature of the water 
decreases,for its heat decreases and the temperature of the tube increases, as its heat increases.) 
fig 3 : Application of the parallel strategy 
AcrEs DE COLING-92, NAgrp.s, 23-28 AO~r 1992 6 1 9 Paoc. OF COLING-92, NANTES, Auo. 23-28. 1992 
tion) andthus is governed by both general rhetorical princi- 
ples and domain dependent considerations. It proceeds as 
an incremental process and each step determines more pre- 
cisely the relevant information. First the notion of theme- 
object, similar to global focus \[Grosz 86\] (and adapted to 
generation by Mc Keown), is used to carry out a rough se- 
lectiou of relevant causal interactions: events which are not 
in (direct or indirect) causal interaction with any theme- 
object are ignored. The theme-object identification de- 
pends on the user interaction mode with the system. In the 
simplest cases, theme--objects can be identified with privi- 
leged properties of the physical system, such as outputs of 
an artifact. But in a dialogue system, an analysis of user 
queries is necessary. 
Second, the selection is refined and emnmunicative acts are 
selected according to user's beliefs and expectations ex- 
pressed in his request. Esr, ccially, communicative acts on 
causal opposition relations are introduced in two kind of 
situation: (a) if a complete description of causal behaviour 
is required or (b) within a concessive strategy, which seem s 
appropriate when the user focuses on an expected but yet 
unoeccurred event. For example, in order to reply to the 
question "why doesn't the mercury level drop ?", causal op- 
position relations are not ignored for it suggests that the 
user is interested with inhibited influences which might 
have caused a mercury drop. Similar considerations are 
expressed in Mc Coy's shemata for correcting misconcep- 
tions about object classifications \[McCoy 89\]. 
IV- TEXT ORGANIZATION 
Text organization deals with multiple communicative acts 
simultaneously. This is one of the main features of our ap- 
proach for it provides the ability to impose a global struc- 
ture on the text. Global coherence cannot be ensured if 
communicative acts are considered independently of one 
another. Additionally, there is not a one to one correspon- 
dence between communicative acts and sentences of the 
text. A single utteranceoften realizes multiplecomm unica- 
live acts and inversely a single act may be spread on several 
utterances. For example, sentence (1) may also be uttered 
so as to achieve the three following acts: informing the 
hearer of mercury expansion and mercury level rise, and 
that these events are causally related. This conception in 
language generation has been initiated by Appelt in the 
KAMP system \[Appelt 85\]. He showed how multiple illo- 
cutionary acts can be achieved in a single utterance. 
Communicative acts are not directly mapped to linguistic 
forms. The textual structure is introduced as an intermedi- 
ary level of abstraction between the specified content and 
the text. It is mainly composed of textual units, inter-clau- 
sal relations, which indicate how these units should be re- 
lated, and order prescriptions. 
IE.I - Inter-Clausal Relations 
Several text generation systems use inter-clausal relations 
which have been introduced in textual linguistics under 
various forms \[Grimes 75\], \[Van Djik 81\], \[Halliday 761. 
R hetorical Structure Theory (RST) \[Mann &Thompson 87\] 
is a signi fican t approach of this trend and it has been partial- 
ly used in text planning systems \[Hovy 88\], \[Moore & 
Swartout 89\]. In RST, the structure of a text is obtained by 
composition of rhetorical relations expressing various 
kindsof dependencies, ~metimes semantic ("subject mat- 
ter" relations in RST terminology), sometimes purely rhe- 
toricai (presentational relation in RST). In contrast, in our 
approach the intended rhetorical effects are considered as 
communicative goals and are specified at the content speci- 
fication level. As a consequence, inter-clausal relations in- 
volved in our system are exclusively semantic. Further- 
more, they are characterized in a way similar to the 
paratactic/hypotactic distinction of rhetorical predicates 
\[Grimes 75\] so that they can take different communicative 
forms. Under the equiponderant form, the same interest is 
assigned to the arguments of the relation while under the 
preponderant form, one of them is presented as more cen- 
tral. 
Arguments of relations are composed of textual units, 
which are semantic translations of conceptual entities. An 
argument may be simple (limited to a single textual unit) or 
complex (composed of several textual units). 
At present our work on inter-clausal relations focuses on 
causal and adversative relations (i.e. on relations involved 
in the expression of our causal primitives). 
IV.I.I - Causal Inter-Clausal Relations 
In the cause-effect inter-clausal relation, events denoted in 
the first argument are presented as being the cause of events 
denoted in tile second. According to the communicative 
form (equipondorant or preponderant) of the relation, dif- 
ferent types of communicative acts may be achieved. The 
following elementary transitions (in a simplified form) 
both lead to the intrc~luction of a cause-effect relation (in 
PROLOG notation, arguments in capitalized letters repre- 
sent variables): 
transition(inform_of_a_causal link, 
inform(cause(C,t~ ), 
cause-effect(equi,C,E)). 
transition(explain with a causal link, 
\[explain(C,E), 
cause(C, E)I, 
cau.~'e-effect(pre, C, E)). 
But there is not necessarily a direct correspondence be- 
tween relations of conceptual level and relations of textual 
level, as in the above transitions. Hence, the following tran- 
sition may hold in some context, even if no causal interac- 
tion exists between on the one hand C 1, E 1 and on the other 
hand C2, E2: 
transition(inform of_multiple_causal links, 
linform(cause(C1 ,El)), 
ir(orm( cause( C2,E2 ) ) l, 
cause-effect( equi,\[ C1 ,C2\],\[E1 ,E21) 
ACTes DE COTING-92, NANTES. 23-28 ^dirt 1992 6 2 0 PROC. OF COLING-92, NAN'rEs, AUa. 23-28, 1992 
IV.I.2 Adversative lnter~7.1ausal Relations 
As mentioned in section H, behavioural description of a 
physical system contains not only causal links between 
events but also causal opposition relations. These relations 
are often phrased in natural language with concessive con- 
nectives (but, though, however,...) which, following Piaget 
\[Piaget 78\], express a discordance between cause am1 ef- 
fect. The sentences (3) and (4) illustrate respectively ex- 
pressions of contrary~cau.~e and opposite-effects rela- 
tions: 
(3) Le tube se dilate uu~is le niveau de mercure raonte. 
(The tube expands but the mercury level rises.) 
(4) Le tube se dilate raais le mercure se dilate aussi. 
(The tube expands but the mercury expands too.) 
The study of linguistic connectives within the linguistic 
Theory of Argumentation \[An~ombre&Ducrot 83\], \[Aas- 
combre 84\] and especially the French connective mais 
(but) has showed that two kinds of semantic opposition 
may be expressed by concessive connectives: direct oppo- 
sition, where one of the two linked propositions is pres- 
ented as an argument for a conclusion oppo~d to the other 
proposition, and indirect opposition, where the two propo- 
sitions are presented as arguments for opposite conclu- 
sions. (3) conveys a direct opposition since the conclusion 
(the mercury level drop) aimed by the first proposition is 
opposed to the second one (the mercury level rise) and (4) 
conveys an indirect opposition since the two propositions 
aim at opposite conclnsions (mercury level drop and mer- 
cury level rise). 
In order to represent these semantic oppositions, we have 
introduced the adversative inter~zlausal relations direct- 
opposition and indirect-opposition. Acts of informing on 
causal opposition relations are then realized by means of 
these adversative relations: 
transition(inforra_of a_contrary_cause, 
inform( contr ar y-catt~e( C,E ) ), 
direct-opposition(pre,C,E)). 
transition(inform_of_oppositeeffects, 
inform(opposite-effects(C1 ,C2)), 
indirect.~opposition(equi,Cl ,C2)). 
But also concessive acts, taking advantage of the conces- 
sive nature of adversative relations: 
transition(concede a contrary_cause, 
lconeede(C1), 
inform( cause( C2,E2 ) ), 
opposite effects(C1 ,C2)1, 
\[indirect--opposition(equi,Cl ,C2), 
cause-eff ect( equi,C2,E2 ) \] ). 
IV.2 - I: rom content specification to textual structure 
A s we diseussed earlier, corn manicafi ve acts should not be 
examined independently ofoneanother if one wants topro~ 
duce well-structured texts. A local transition represents a 
possible treatment of a limited number of communicative 
acts at textual level. Hence, choice and application of local 
transitions are governed by discourse strategies, They de- 
termine transitions which may be applied after an analysis 
of the overall set of communicative acts. In particular, they 
cxploit the underlying conceptual structure ofcomnurnica- 
live acids. For cau~l explanations, the mainly used strate- 
gies are the following: 
• Causal Chain Strategy: If the underlying conceptual 
structure is a causal chain and communicative acts are of 
the inform or explain type then lollow causal order and 
apply transitions informofa_causallink and 
explain with a causal link (cffig 2). 
* Parallel Strategy: It is a form of parallelism rhetorical 
"figure" which may be used when the underlying con- 
ceptual structure is comlx~sed of two causal chains with 
a common initial cause. This strategy also exploit the 
causal order and transitions inform of common 
cause link and inform of multiple causal_links hold 
the highest priorities (cffig 3). 
tramition(inform of common_cause_linLL 
\[inform(cause(C,El )), 
inform( cause( C ,E2 ) ) \] , 
cause-effect( equi,C ,\[E l ,E21) 
• Concessive Strategy: This strategy deals with concessive 
acts which involve causal opposition relations. 
A preferential order is suggested: the conceded fact pre-- 
cedes the other units. 
Priorities are assigned to the strategies so that conflicts can 
be solved. 
IVJ - Textual Structure Segmentation 
The purpose of the segmentation task is to delimit the con- 
tent of the ~utences which will constitute the final text. 
The determination of a sentential content involves several 
heuristics. Some of them aim at increasing the scope of the 
sentence while others aim at reducing it. One of the main 
heuristics deals with the communicative nature of inter- 
clausal relations: since preponderant relations often appear 
in subordinate forms at syntactic level, equiponderant rela- 
tions are privileged. Hence, delimitation of sentential con- 
tent starts with the choice of equiponderant relations. The 
content is then completed with preponderant relations. In 
addition, the number of textual units of a sentence is limited 
and introduced relations may be removed from the senten- 
tial content if it contains too many textual units. The seg- 
mentation is "also coustralued by the conceptual nature of 
inlormation units. For instance, an initial cause may be 
realized in a single sentence. 
AcrEs DE COLING-92, NAIVrEs, 23-28 APt)l" 1992 6 2 1 Prtoc. ov COLING-92, NANTES, AUO. 23-28, 1992 
V- SURFACE GENERATION VI - RELATED WORKS 
The text planner is independent of the surface generation 
component. (except the stylistic adjustment task). Two dif- 
ferent types of sentence generation approaches have been 
examined. 
Oar main approach aims at coupling the text planning sys- 
tem with a generic sentence generator \[Fournier 91 \] based 
on the Meaning-Text Theory \[Mel'cuk 88\]. In the Mea- 
ning-Text Model, the input representation is a semantic 
network enriched with several structures, mainly the bipar- 
titions theme/rheme, asserted/presupposed and given/new. 
The construction of semantic networks is based on a con- 
ceptual-semantic dictionary, which specifies correspon- 
dences between concepts and semantic definitions. The gi- 
ven~new structure has already been integrated in text 
generation systems (e.g. INirenburg 88\]). It denotes speci f- 
ic linguistic constructions and will constrain choices dur- 
ing realization. For instance, a predicative structure will be 
rather realized as an assertion if it belongs to the new part 
or as a nominalization if it belongs to the given part. This 
characterization of semantic representations plays a crucial 
role from a textual perspective. The generation process 
must be sensitive to progression of new information 
through the text, even at the realization stage. In this spirit, 
distribution of semantic units among given and new parts 
is carried out with regard to conceptual units introduced in 
previous sentences. Predicative structures of the semantic 
network whose referents have been previously evoked are 
systematically marked as given and the others as new. Once 
the semantic representation is built (only the semantic net- 
work and the given/new structure have been considered), 
the realizer produces surface forms and paraphrases may be 
generated, essentially by lexical substitutions. 
The second approach we are experimenting relies on equiv- 
alence classes between sentence structures, which may be 
compared to Discourse Grammars described in \[Danlos 
85\]. An equivalence class represents potential realizations 
of a kind of semantic content. Our investigation is currently 
limited to classes related to causality. We have defined 
classes for expressing causal and opposition relations. This 
categorization of linguistic expressions provides a suitable 
basis for stylistic adjustment mechanisms. Syntactic and 
lexical constraints arising from stylistic considerations 
govern the choice of sentence structures within classes. In 
particular, we have implemented a "principle of stylistic 
variety" which prevents excessive use of similar syntactic 
constructions. This principle is particularly useful in causal 
explanation generation because concepts involved are of 
similar types. To illustrate this point, let us consider the text 
in figure 2. When generating the last sentence, stylistic ad- 
justment recommends the realizer to avoid coordinative 
structures (with connectives such as thus, consequently, ...) 
since such a construction has been used in the previous sen- 
-tence. A structure with a verb conveying causality (e.g. to 
cause, to result in, ...) is then selected. 
Our system benefits from a number of earlier works on lan- 
guage generation, as those we have mentioned along the 
above sections. We will focus the comparison on two par- 
ticular points. The first concems the overall architecture 
and modularization in text generation systems; the second 
deals with text planning applied to causal explanation gen- 
eration. Text planning and surface generation have often 
been conceived as separated components. But separation of 
tasks is not always extended within text planning. For ex- 
ample, rhetorical schemata used in TEXT \[McKeown 85\] 
specify in the same structure not only information units 
which may be introduced bat also their orders in the final 
text. Other systems as GOSSIP \[Carcagno & Iordanskaja 
89\] and PAULINE \[Hovy 88b\] have isolated content deter- 
mination from text structuring. The approach described in 
this paper also proposes a clear separation of the content 
specification and the textual organization skills. The main 
motivation behind this division is that it is convenient to de- 
cide what to say (more precisely, to set (almost) all commu- 
nicative goals to achieve) before taking any organizational 
decisions. Hence, content specification proceeds regard- 
less of structuring considerations. However, we think that 
a more cooperative interaction between these two major 
components is necessary to allow goal revisions at the text 
organization level. 
Oar system can also be compared to TAILOR \[Paris 87\] 
which generates natural language descriptions of physical 
devices. A significant advantage of TAILOR is its ability to 
combine structural and causal description strategies. Nev- 
ertheless, causal interactions are restricted to simple causal 
links and there is no attempt to explicit the roles they can 
play la discourse. 
VII - FUTHER DEVELOPMENTS 
First a better conceptual characterization of physical sys- 
tems would contribute to improve the quality of causal ex- 
planations. We need a more precise description of causal in- 
teractions which allows, for instance, to discern enable~ 
ment conditions from other causal links. 
With regards to text planning, number of extensions are 
possil)le. We intend to define strategies for structural de- 
scriptions and also enhance the control mechanisms of dis- 
course strategies. Furthermore, practical validation of the 
overall approach requires a larger coverage of communica- 
tive acts. Another interesting extension would consist in 
relating stylistic adjustment mechanisms to pragmatic fea- 
tures \[Hovy 89\] in order to strengthen context sensitivity. 
Acknowledgments 
I would like to thank Corinne Fournier and Marie-42hris- 
tine Escalier for their comments on earlier versions of this 
paper. 
ACRES nE COLING-92, NANTES. 23-28 Ao0"r 1992 6 2 2 PROC. OF COLING-92, NANTES, AUG. 23-28, 1992 

REFERENCES 

- Anscombre, J.C. "Grammaire traditionnelle et gram- 
maire argumentative de la concession", Revue internation- 
ale de philosophie, 155, 1984. 

- Anscombre, J.C. & Ducrot, O. L'Argumentation dons la 
langue, Bruxelles, Mardaga, 1983. 

- Appelt, D.E. "Planning Natural Language Referring Ex- 
pressions", Artificial intelligence, 25, 1985. 

- Carcagno, D. & lordanskaja, L. "Content Determination 
and Text Structuring in GOSS IP",2nd European Workshop 
on Natural Language Generation, 1989. 

- Cerbah, E & Fournier, C. & Raccah, P.Y. "Qualitative 
reasoning and Argumentation: A study of some affinities 
when Generating Causal Explanations", 1st Workshop on 
Qualitative Reasoning and Decision Support Systems, 
1991. 

- Cohen, P. & Perranlt, C.R. "Elements of a Plan Based 
Theory of Speech Acts", Readings In Natural Language 
Processing, Morgan Kaufman Publishers, 1986. 

- Danlos, L. Gdndration automatique de textes en langues 
naturelles, Masson, 1985. 

- De Kleer, J. & Brown, J.S. "A Qualitative Physics Based 
on Confluences", Artificial Intelligence, 24, 1984. 

-Elhadad, M. & Mc Keown, K. "Generating Connectives", 
COLING 90. 

- Forbus, K.D. and Gentaer, D. "Causal Reasoning about 
Quantifies", Proceedings of the Fifth Annual Conference 
of the Cognitive Science Society, Lawrence Erlbaum Asso- 
ciates, Inc, 1983. 

- Forbus, K.D. "Qualitative Process Theory", Artificialln- 
telligence, 24, 1984. 

- Forbus, K.D. "Qualitative Physics: Past, Present, and Fu- 
tare", Exploring Artificial intelligence, Howard Shrobe 
(eds). Morgan Kaufmann Publishers, Inc, 1988. 

- Foumier, C. "Un g6ndrateur de textes fond6 sur le modSle 
Sens-Texle",Technical ReportDassaultAviation, 1991. 

-Grimes, J.E. The Thread of Discourse. Mouton, 1975. 

- Grosz, B.J. & Sidner, C. "Attention, Intention and the 
StructureofDiscourse",Computational Linguistics, 1986. 

- Halliday, M.A.K. & Hasan, R. Cohesion in English, Lon- 
don, Longman, 1976. 

- Hovy, E.H. "Planning Coherent Multisentential Text", 
26thACL, 1988. 

- Hovy, E.H. "Pragmatics and Natural Language Genera- 
tion", Artificial intelligence, 43, 1989. 

- J6z6quel, P. & Zimmer, L. "SQUALE : manuel d'utilisa- 
tion", Technical Report Dassault Aviation, 1992. 

- Mann, W.C. & Thompson, S.A. "Rhetorical Structure 
Theory: A Theory of Text Organization", ISI/RS-87-19l), 
1987. 

- McCoy, K.F. "Generating Context-Sensitive Responses 
to Object-Related Misconceptions", Artificial Intelli- 
gence, 41, 1989. 

- Mc Keown, K. Text Generation, New York, Cambridge U. 
Press, 1985. 

McKeown, K. & Swartout, W."Language generation and 
explanation", Advances inNatural Language Generation. 
Zock and Sabah (eds), London, Pinter, 1988. 

Mel'cuk, I. Dependency Syntax: Theory and Practice, 
SUNY, 1988. 

- Moore, J. & Swartout, W. "A Reactive Approach to Ex- 
planation", I.ICA189. 

- Nirenburg, S. & Nirenburg, I. "A framework for lexical 
selection in natural language generation", COLING 88. 

- Paris, C.L. "Combining Discourse Strategies to Generate 
Dcscriptions to Users Along a Naive/Expert Spectrum",lJ- 
CA187. 

- Piaget, J. Le Jugement et le raisonnement chez l'enfant, 
Delactmux et Niesfl6, 1978. 

- Suthers, D.D. "Perspectives in Exphmation", COINS 
Technical Report 89-24, 1989. 

- van Dijk, T.A. Studies in the Pragmatics of Discourse, 
Mouton, 1981. 
