Compound Nouns in a Unification-Based MT System 
Pierrette Bouillon Katharina Boesefeldt 
ISSCO, 54 route des Acacias 
1227 Geneva, Switzerland 
pb~divsun.unlge.ch 
Graham Russell 
Abstract 
This paper describes an approach to the 
treatment of nominal compounds in a ma- 
chine translation project employing a modern 
unification-based system. 
General problems connected with the analy- 
sis of compounds are briefly reviewed, and the 
project, for the automatic translation of Swiss 
avalanche bulletins, is introduced. Avalanche 
bulletins deal with a limited semantic domain 
and employ a sublanguage in which nominal 
compounds occur frequently. These and other 
properties of the texts affect the treatment of 
compounds, permitting certain simplifications, 
while leaving a number of possible alternative 
analyses. 
We discuss the different problems involving 
the translation of compounds between German 
and French, and show how the computational 
environment in use permits two different ap- 
proaches to the problem: an interlingua-based 
approach and a transfer-based approach. Fi- 
nally, wc evaluate these approaches with re- 
spect to linguistic and computational consider- 
ations applicable in a MT-system dealing with 
a limited semantic domain and describe the so- 
lution that has actually been implemented. 
1 Compound Nouns 
Compound words pose well-known problems for linguis- 
tic description in general, and some additional ones for 
natural language processing in particular: 
Identification: how can compounds be distinguished 
from other words and phrases? 
Segmentation: what are the components of a com- 
pound? In many languages, including German, ortho- 
graphic convention is such that compounds are writ- 
ten as single units. I 
Disambiguation: what is the correct analysis of a com- 
pound? On the widespread assumption that com- 
pounds have a recursive binary structure, any occur- 
rences with more than two basic elements will ad- 
1Elements of German compounds may however be sepa- 
rated by the so-called "Fugenseichen" (8, en, etc.). 
mit multiple analyses, 2 from which, normally, a single 
candidate must be selected. 
Interpretation: how can the meaning of a compound 
word be derived from the meanings of its parts? For 
many purposes, there is little point in performing any 
of the other tasks unless this is feasible. 
It is clear that solutions to these four problems may be 
closely interrelated; ill-formed interpretations may per- 
mit unwanted analyses to be filtered out, the correct 
analysis will constrain possible segmentations, and so 
on. 
In what follows, we outline an approach to the treat- 
ment of compounds within a specific limited application, 
the automatic translation of Swiss avalanche warning 
bulletins, which exploits the nature of the texts involved 
in order to translate compounds efficiently and correctly. 
We first give a brief overview of the project, paying 
special attention to phenomena related to compounds 
and their translation. We then describe ELU, 3 the soft- 
ware employed for translation, and discuss a number of 
the different treatments that it permits, motivating our 
choice of analysis with illustrations of their weak and 
strong points. 
2 Compounds in Avalanche Bulletins 
The avalanche bulletins are issued by IFENA (the Fed- 
eral Institute for the Study of Snow and Avalanches) 4 a 
number of times a week during the winter season, the ex- 
act frequency of their appearance depending on weather 
conditions. Bulletins are prepared in German, and are 
translated into the other official Swiss languages, French 
and Italian, before publication. The current state of af- 
fairs, in which the source language is exclusively Ger- 
man, may change in future, and for this reason it has 
been decided to implement a reversible system (Bouillon 
and Boesefeldt, 1991a; Bouillon and Boesefeldt, 1991b). 
Avalanche bulletins describe a fixed and specifiable 
semantic domain, and employ a language restricted in 
both vocabulary and syntactic variety. They contain a 
large number of compounds, with differing grammatical 
2See Church and Patil (1982: p. 140ff.) for discussion. 
3"Envizonnement Linguistique d'Unification" (Estival, 
1990). See also Johnson and Rosner (1989) for a description 
of the earlier UD system on which ET.u is based. 
4The project is partially supported by IFENA. 
  
 209 
properties. It is the interpretation of compounds, to- 
gether with the closely related issue of structural disam- 
biguation, which most researchers have addressed (Finin, 
1980; Isabelle, 1984; Sparck Jones, 1983). In an ap- 
plication such as avalanche bulletin translation, how- 
ever, the necessity for a deep or sophisticated analy- 
sis of compound meanings is not obvious. On the one 
hand, the number of compounds appearing in the sub- 
language is highly restricted (approximately 400); they 
may therefore be listed exhaustively, and the question 
of disambiguation does not arise. And on the other, 
the interpretation of a compound is given by its estab- 
lished translation - the meaning of Lawinengefahr ('dan- 
ger of avalanches') is just the information necessary to 
produce the required target language expression danger 
d'avalanches (or pericolo di valanghe, etc.). 
In the current system, the main difficulty thus consists 
of establishing a systematic and general relation between 
single words in German and complex nominal structures 
in French, without defining the semantic and pragmatic 
roles of the different parts of the compound. 
2.1 Variation in German Compounds 
A purely categorial analysis of the German compound 
nouns contained in the avalanche warning bulletins re- 
veals the following internal combinations of nouns, ad- 
jectives, verbs and prepositions: 
\[ADJ N\] Neuschnee 
'new snow' 
\[N N\] Lawinengefahr 
'danger of avalanches' \[V N\] 
Schwimmschnee 
'floating snow' IN \[ADJ N\]\] 
Alpensgdhang 
'southern slope of the Alps' 
\[\[ADJ N\] r~\] geuschneezuwachs 
'increase in new snow' \[Iv N\] N\] 
Schwimmsehneebildung 
'development of floating snow' 
\[\[N N\] N\] Schneebrettgefahr 
'danger of snow patches' 
\[\[PREP N\] N\] Uberschneesprengung 
'explosion above the snow' 
The 'head' of a compound is its final constituent 
which determines its properties (in the examples above, 
-schnee, -gefahr, -zuwachs, etc.), and may itself be a com- 
pound (-sgdhang, for example). Constituents of a com- 
pound are not inflected, although, naturally, the com- 
pound as a whole may be. 
There is a very strong tendency to restrict 
the length of compounds to three elementary con- 
stituents; of approximately 400 compound nouns in 
the bulletins, issued over the past few years, only 
one (Naflschneelawinengefahr: 'danger of wet-snow 
avalanches') contains four. s Complex meanings that 
SThis distribution can also be found in other domains 
(Boesefeldt, 1989) as well as in general language (Fleischer, 
1982). 
might be expressed by a fourth element within a com- 
pound generally occur in the form of a noun taking the 
compound as complement (e.g. Abgang yon Naflschnee- 
lawinen, literally: 'going-down of wet-snow avalanches')~ 
or conventional adjectival modification (e.g. sehr gross~ 
allgemeine Schneebre~tgefahr 'very great general dange~ 
of snow patches'). 
2.2 The French Translation 
Compound nouns in the German original texts are usu- 
ally translated into French as complex nominal phrase., 
containing adjectival or prepositional subparts (e.g 
Neuschnee: neige frafche; Schneebre~gefahr: danger d~ 
plaques de neige). The details of translation differ in 
number of aspects. First of all, the French phrases do not 
always contain the same syntactic categories as the cot. 
responding German compounds. The noun-noun com- 
pound Oberflachenschicht ('surface layer'), for instance 
is generally translated by the sequence of noun and post. 
nominal adjective couches superficielles. 
Secondly, prepositions within French translations o: 
German compounds vary from case to case: Schnee 
brettgefahr is translated as danger de plaques de neige 
Schneeraster as grille d_ neige and Lawinenforschung a,. 
recherche su_.._Er les avalanches. The use of articles withix 
these PPs also varies: tempdra~ure de la neige ('snov 
temperature') has the definite article, while quan~i~d d~ 
neige ('snow quantity') has none. Even though the us~ 
of the different prepositions and articles might contaii 
clues concerning the possible semantic and pragmatic re. 
lation between the different elements of the compound, 
(d denoting an instrument, an omitted article an abstracl 
or partitive, etc.), the size of the corpus (only about 100( 
words including nouns, verbs, articles etc.) makes it im 
possible to draw any general conclusion which could bq 
implemented. 
Finally, certain parts of German compounds arq 
systematically omitted from the corresponding Frencl 
phrase: Schneebret~anrifl ('rupture of a snow patch' 
translates as rupture de plaque, where the word neig 
does not appear. 
3 The treatment of compounds with 
ELU 
Since 1990, the ELU system has been used at ISSCO fo 
the implementation of a machine translation system fo 
Swiss avalanche warning bulletins (Boesefeldt and Bouil 
lon, 1991; Bouillon and Boesefeldt, 1991a; Bouillon ant 
Boesefeldt, 1991b). ELU is a unification-based gram 
mar development environment in the style of PATR-I 
(Shieber, 1986), designed especially for experimentatio\] 
with machine translation. 
ELU is composed of three modules: a parser, a gen 
erator and a reversible transfer module. Since it is th, 
last of these that chiefly concerns us here, we shall sa: 
nothing about the parser or generator. ELU support 
various methods of translation. By writing grammar 
that share meaning representations, one may implemen 
an interlingua-based system; the result of analysing 
source-language text is used directly as the basis for gen 
  
 210 
erating the target-language text, and transfer is unneces- 
sary. At the other extreme, ELU transfer rules are capa- 
ble of performing arbitrary transformations on an input 
structure, and thus permit whatever variety of transfer 
the user desires. ° A transfer-based approach to transla- 
tion has been selected for the current project. 
Transfer rules are written which state binary relations 
over the representations of texts from the source and 
target languages so as to associate the analysis of one 
language with the synthesis of another. There are two 
kinds of transfer rule, treating atomic and complex fea- 
ture structures. For example, the following two FSs, rep- 
resenting the German phrase die Ge/ahr and its French 
equivalent le danger, 
DETYPB definite L DBTYPE definite 
are related by means of the following four rules: 
:TA: gefahr danger 
:TA: definite definite 
:T: pred :T: detype 
:LI: <* pred>= X :LI: <* detype> = X 
:L2: <* prod> = Y :L2: <* detype> = Y 
:X: X = Y :X: X = Y 
The transfer rules establish a relation between the 
pairs of atoms gefahr - danger and definite - defi- 
nite as the values of the attributes PRED and DETYPZ in 
the two representations. The rules pred and detype ab- 
stract away from the particular values which their paths 
may take; a statement of the form ":g: g = Y" indi- 
cates that the success of the rule in which it appears is 
conditional on the success of other rules involving what- 
ever structures are bound to X and Y during the transfer 
process. Transfer of a FS thus proceeds recursively, be- 
ginning with the root, and, provided that no failure oc- 
curs, terminating with the atoms which form the 'leaves' 
of the FS. For further details, see Estival et ai. (1990) 
and Russell etal. (1991). 
Two broad classes of solutions suggest themselves for 
the treatment of compounds in ELU: one, interlingua- 
oriented (cf. Section 3.2), in which the representations in 
the two languages are broadly similar, as in the Gefahr 
- danger example, and the other exploiting more fully 
the transfer mechanism (cf. Section 3.1). In the former 
case, it is possible to pass from a German compound to 
a complex French nominal phrase by employing general 
transfer rules of the kind required by other aspects of 
the translation task, while in the latter, different repre- 
sentations are related by complex transfer rules specific 
to the treatment of compounds. 
3.1 The transfer-based approach 
At first view, the transfer-based approach seems to be 
the more natural. 
O,,Codescription, analyses of the type proposed by Kaplan 
et al. (1989) are also possible. 
In German, compounds can be introduced in the lexi- 
con as simple words and treated in the same way by the 
grammar. The NP eine Schneebrettgefahr, for example, 
will be represented by a feature structure similar to that 
provided for the simple eine Gefahr. The semantics of 
adjectival modifiers is encoded as elements in a list, since 
the number of these may vary. Lists are indicated in 
attribute-value diagrams by '(...)' and in transfer rules 
by '\[... \] '; those shown here are empty: 
PRED schneebrettgefahx" 
DETYPE indefinite 
~oD ( > 
PRED gefahr \] 
DETYPE indefinite 
MOD < } 
The French grammar constructs the translation of a 
German compound from individual words, the head of 
the phrase being lexically specified for its complement 
and/or modifiers. For example, danger subcategorizes 
for a PP headed by the preposition de, and contain- 
ing a NP which lacks an article and whose head noun 
is one of a class that includes plaques. The represen- 
tation constructed for the French equivalent danger de 
plaques de neige will be considerably more complex than 
the German, with sub-FSs relating to the head (danger) 
and complement, which itself will be a complex struc- 
ture with information concerning a head (plaques) and 
complement (neige). Since the prepositions in these con- 
structions are determined by the lexieal properties of the 
nouns, they convey no information in these eases, and do 
not appear in the representation, the French grammar 
accounts for their correct distribution. 
r nei e\] 
ARGS I DETYPE non \[ 
LMOD <} J 
ARGS I PILED plaques 
I DETYPE non 
LMOD < > 
PRED danger 
DETYPE indefinite 
~ov ( ) 
This technique makes it necessary to write a trans- 
fer rule for every compound in order to pass from the 
German compound to the different French words: 
:T: 
:LI: 
:L2: 
:I: 
Schneebrettgefahr 
<* pred>= schneebrettgefahr 
<* args args pred>= neige 
<* args args detype> = non 
<* axgs axgs mod> = \[\] 
<* args pred>= plaques 
<* args detype> = non 
<* args rood> = \[\] 
<* prod> = danger 
It presents a number of disadvantages. In the first place, 
certain generalizations which we consider important can- 
not be taken into account. For example, an element 
may appear in a number of different compounds, and 
the same information has to be repeated in each case. 
  
 211 
For every compound including Gebiet ('region'), for ex- 
ample, we have to repeat the fact that the noun Gebie~ 
translates as rdgion: 
:T: 
:LI: 
:L2: 
Simplongebiet 
<* pred> = simplongebiet 
<* args pred> = simplon 
<* args detype> = non 
<* args mod> = \[\] 
<* pred> = region 
:T: 
:LI: 
:L2: 
Engadingebiet 
<* pred> = enEadinEebiet 
<* ares pred> = enEadine 
<* ares detype> = non 
<* ares mod> = \[\] 
<* pred> = region 
Moreover, as this term also exists as a simple word, we 
also have to add a transfer rule to translate the simple 
word: 
:TA: gebiet region 
No account is taken of the fact that the elements in- 
side the compounds correspond to isolated words and 
translate in the same way; moreover, a larger number of 
transfer rules is required, which, in turn, decreases the 
efficiency of the system. 
The use of this technique leads to an additional prob- 
lem. It is possible to add an adjective to a com- 
pound which already contains an adjective, e.g. ganze 
AIpensCzdhang ('whole southern slope of the Alps'). This 
phrase has the following representation: 
I PRBD alpensudhang 1 
MOD ( g~z ) J 
The representation of equivalent French expression ~out 
le versant sud des Alpes, however, is as follows: 
\[P. D .Jpes \] 
ARGS I DETYPE definite 
L MOD ( ) 
I PRED vezsant 
L MOD ( tout, sud ) 
For the transfer between the two languages it is neces- 
sary to establish a relation between the dements inside 
the lists. We cannot be sure, however, that the adjective 
which is a part of the German compound will always 
be at the same place inside the list in French. For this 
reason, it was decided to distinguish within the French 
description between adjectives which compose in Ger- 
man and adjectives which do not. This distinction which 
might seem undesirable at first sight has been proved sat- 
isfactory and generalizable for the sublanguage we treat 
and could be quite easily established. 
An adjective cannot be found inside a German com- 
pound if it is part of a coordination, if it is modified by 
an adverb, if it has a particular morphological suffix or 
if it is a participle which is used adjectivally, if it has 
been given a certain semantic type 7 (adjectives qualify- 
ing snow always compose, adjectives expressing a degree 
of danger never compose), if there already is another ad- 
jective inside the compound or if the compound contains 
more than three elements. In all these cases it will there- 
fore become the value of the path MOD : NOM and not 
the value of the new path MOD : N_COMP. 
The representation of the nominal phrase ~out le ver- 
san~ sud des AIpes is then modified as follows: 
I  pes \] 
ARGS |DETYPE defmlte| 
L MOD ( ) J ~ 
PRED versant 
l MOD \[NOM tout 1 LN_CO~ sud J 
The attribute MOD has thus been subdivided into two 
parts, MOP : NOM for adjectives like ~ou~ that do not 
compose and MOD : N_COMP for adjectives like sud that 
do compose. 
The transfer rules for the transfer between ganze 
AIpens~dhang and the French ~o~ le versant sud des 
Alpes can then be simplified and made more general: 
: T : Alpensudhan E 
:LI: <* pred> = alpensudhan E 
:L2: <* ares pred> = alpes 
<* ares detype> = non 
<* ares mod> = \[\] 
<* mod n comp> = sud 
<* prect> = versant 
:l: - 
:T: nom 
:LI: <* mod nom> = X1 
:L2: <* mod nom> = X2 
:X: Xl = X2 
:Ta: ganz tout 
This technique has been successfully implemented and 
was the first step towards the second solution to be pre- 
sented here, the interlingua-oriented solution. 
3.2 An interllngua-orlented approach 
In order to obtain the same representation in German 
and in French, the German compounds have to be as- 
signed in the lexicon a complex representation in which 
the word is semantically decomposed into head and com- 
plement(s). Moreover, certain information concerning 
the internal structure of the French compounds has to 
be added to the German representation during the anal- 
ysis. This process can be clone quite easily in ELU by 
means of macros, s The desired structures need only to 
be defined once, the adequate values being instantiated 
at the place of the variables. 
rLexical items are typed according to the contexts in 
which they may appear. 
SET.u macros resemble a more powerful version of PATR- 
II 'templates', permitting the use of arguments, and multiple 
or recursive definitions. 
  
 212 
The NP Alpens~dhang will thus receive the same rep- 
resentation as its French equivalent: 
PRED alpen \] 
definite ~IIG S \[ DETYPB I \] 
L MOp \[ N_COMP ( LNoM ( 
PRED hang 
\[ PRED a\]pes \] 
definite 
I\] \[Mo  < LNOM ( 
PRED versant 
The paths ARGS : MOD : NOM and ARGS : DZTYPE in the 
German representation have been added because they 
are added by grammar rules constituting French NPs 
and PPs. 
This technique makes it possible to simplify the trans- 
fer rules, which can then be used for the transfer of not 
only compounds, but also other items: 
: T : ncomp 
:LI: <* mod n_comp> = X1 
:L2: <* mod n_comp> = X2 
:X: Xl = X2 
:T: nom 
:LI: <* mod nom> = X1 
:L2: <* mod nom> = X2 
:X: X1 = X2 
:T: pred 
:LI: <* pred> = X1 
:L2: <* pred> = X2 
:X: X1 = X2 
:T: args 
:LI: <* args> = X1 
:L2: <* args> = X2 
:X: X1 = X2 
:T: detype 
:LI: <* detype> = Xl 
:L2: <* detype> = X2 
:X: X1 = X2 
:TA: alpen alpes 
:TA: hang versant 
:TA: sud sud 
:TA: definite definite 
The semantic decomposition of the German com- 
pounds not only decreases the number of transfer rules, 
but a/so makes the transfer more general and coherent, 
as the different elements of a compound are translated 
as simple words. 
The first problem of this technique consists of deciding 
what complex semantic representation to give German 
compounds. The decomposition could be performed ac- 
cording to the categories in either German or the target 
language. A disadvantage of the former is that, for the 
reasons given in section 2.2, transfer rules for irregular 
cases must be added. The latter approach avoids this 
difficulty by assigning the noun-noun compound Son- 
nenlagen ('sunny places'), for example, a representation 
in which the contribution of the element sonnen- and 
-lage are classified as modifier and predicate respectively, 
rather than predicate and argument: 
PRED lage \] 
MOD \[ NOM < ) l 
L N_COMP ( sonnig ) J 
This representation corresponds to the representation for 
the French equivalent, endroit ensoleilld: 
I PRED endroit \] \[.oM 
< > 1 LN-COMP ( ensoleil\]~ ) \] 
As sonnig is not the value of the path MOD : NOM, this 
representation does not interfere with the representation 
of the possible nominal phrase sonnige Lage: \[._ ,.,. 
MOD \[NOM < so~ig 
LN-COMP < ) 
But there is another reason to treat the irregulari- 
ties between German and French in the German lexicon. 
Two different words can also be used as synonyms in- 
side and outside of a compound (e.g. Grisons is trans- 
lated by Graub~nden in German, but Grisons nord is 
translated by Nordb~inden and not by Nordgraub~nden). 
Nordb~nden has therefore been assigned a representation 
suitable for the generation of the French nominal phrase, 
nord des Grisons: 
| DETYPB definite ARGS 
Ll.O D/ < LN-Co~ < 
PRED nord 
/ DETYPE definite 
LMDD F N°M ( \[N_oo~ < 
PRED nord 
This representation for the German compound, however, 
does not seem very satisfactory because it would be nec- 
essary to write a special atomic transfer rule in addition 
to the transfer rule for Graub~nden: 
:TA: bunden grisons 
:TA: graubunden grisons 
  
 213 
Moreover, this would introduce an unwanted ambigu- 
ity when translating from French into German, since the 
French Grisons would now be related to both B~nden 
and Graubgnden. Nordbgnden is therefore assigned a 
representation in which nord and Graubgnden are related 
as shown here: 
\[ PR.BD graubunden \] definite 
/ I\] ( L N-COZ~ < PPd~D nord 
Changes to lexical specifications can be made very easily 
by altering the value of the relevant path in the lexicon 
entry, without affecting the surface form of the lexicai 
item. 
Similarly, certain parts of the German compound have 
to be eliminated from the representation in the case of 
French nominal phrases in which certain parts of the 
German compound have been excluded (cf. section 2.2). 
This technique, however, cannot be applied if a com- 
pound without the element that is missing in French 
exists in German because the same French representa- 
tion would then lead to the generation of two different 
nominal phrases in German. 
In a number of cases the content of a compound can 
also be expressed by a complex nominal phrase with 
only slight stylistic differences (Fleischer, 1982: p. 20- 
21). The semantic representation used for compounds, 
however, is also used to express complex NPs in German; 
these constructions are also translated into French nomi- 
nal phrases. In constrast to the case of adjectives within 
compound, which are assigned a special path in the rep- 
resentation (MOD : N_COMP), a complex NP in German, 
e.g. Abgang yon Lawinen ('going-down of avalanches') 
will get the following representation: 
PRED lawinen \] 
ARGS \[DETYPE indefinlte I\] 
L NCOl~ ( PRBD abgang 
which is the same as the representation for the com- 
pound Lawinenabgang, containing exactly the same in- 
formation. In order to avoid over-generation resulting 
from the same representation being used for two different 
syntactic constructions, the possible syntactic structures 
in German have been restricted. 
Analysis of the corpus of avalanche bulletin texts 
brought to light two types of case in which the same 
representation could be obtained for a compound and 
a complex NP. In the first, a fourth element either ap- 
pears with a compound in a complex NP or has been 
included as an element of the compound itself. As both 
constructions contain exactly the same information, it 
was decided to eliminate the second possibility, and al- 
low no more than three elements within a compound. 
For example, the word Naflschneelawinengefahr ('danger 
of wet snow avalanches') which is sometimes used instead 
of Gefahr yon Naflschneelawinen does not appear in the 
lexicon, and the phrasal version is used instead. 
It was also observed that, according to the degree 
of lexicaiisation, nouns originally derived from a verb 
such as Abgang can form a part of a compound (e.g. 
Neuschneeablagerung - 'deposit of wet snow') as well 
as a part of a nominal phrase (e.g. Verfesiigung der 
Schneedecke - 'consolidation of the snow cover'). As no 
general syntactic rule could be found to deal with this 
phenomenon it was decided to treat it on a case by case 
basis according to the occurrence of derived nouns in the 
avalanche bulletins. 
The restrictions introduced into the German gram- 
mar, which are due to general language tendencies as well 
as to the language used in the avalanche bulletins, thus 
make it possible to avoid over-generation for German 
compounds without introducing an additional path for 
complex NPs. In case of the sublanguage we are treat- 
ing, semantic decomposition therefore does not compli- 
cate the transfer as it is carried out according to the 
requirements of the target language. If phenomena of 
another language which might be added later contradict 
the current decomposition, specific transfer rules can be 
added to the system. 
4 Conclusion 
To conclude, we return to the four problems stated in 
section 1: 
Identification of compounds presents no problem; a 
compound is listed in the lexicon just like any other 
noun, except that its representation is more complex. 
Segmentation of compounds is not performed; the 
number and variety of compounds in the texts is suffi- 
ciently restricted, and the facilities of ELU sufficiently 
flexible, to make morphological analysis redundant. 
Disambiguation of the structure of compounds is not 
necessary, since they are not morphologically decom- 
posed, and since the 
Interpretation of compounds is given by their lexical 
entry in combination with the transfer rules which 
accept meaning representations of sentences in which 
compounds appear. 
In this paper we have outlined our approach to the treat- 
ment of compounds in a machine translation system 
dealing with the limited domain of avalanches in Switzer- 
land. We have shown how we used ELU, a unification- 
based hnguistic environment which has been developed 
for the implementation of machine translation systems 
for the testing of two different approaches to the trans- 
lation of compounds: a transfer-based approach and an 
intedingua-oriented approach. The implementation of 
the these two approaches has shown that even though 
they have both proved satisfactory the interlingua ap- 
proach seems more efficient and more general for the 
treatment of the sublanguage because it permits a rapid 
and coherent reversible transfer. The decomposition oJ 
the German compounds which is indispensible in an in- 
terlingua approach and did not complicate the system 
made it possible for us to solve the problems related tc 
the translation of compounds which are the changing ot 
categories, the use of synonymy and the removal or ad- 
dition of certain elements in the target language. 

References 

Boesefeldt, K. (1989) Le probeme de noms composds en 
allemand, Ecole de Traduction et d'Interpretation, 
University of Geneva. 

Boesefeldt, K. and P. Bouillon (1991) "Le r61e de la 
reprdsentation sdmantique duns un syst~me de tra- 
duction multilingue," Working Paper no. 58, ISSCO, 
Geneva. 

Bouillon, P. and K. Boesefeldt (1991a) "La Traduction 
Automatique des Bulletins d'Avalanches de la Su- 
isse," Colloque International sur l'Environnement 
Ttaductionnel, Mons. 

Bouillon, P. and K. Boesefeldt (1991b) "Applying an ex- 
perimental MT system to a realistic problem," Pro- 
ceedings of Machine Translation Summit III, Wash- 
ington D.C., lst-4th July 1991, 45-49. 

Church, K. and R. Patil (1982) "Coping with Syntactic 
Ambiguity, or How to Put the Block in the Box on 
the Table," Computational Linguistics, 139-149. 

Estival, D. (1990) "ELU User Manual," Technical Report 
1, ISSCO, Geneva. 

Estival, D., A. Ballim, G. Russell and S. Warwick (1990) 
"A Syntax and Semantics for Feature-Structure 
Transfer," Proceedings of the 3rd International Conference on Theoretical and Methodological Issues in 
Machine Translation of Natural Language, Austin, 
Texas, 11th-13th June 1990, 131-143. 

Finin, T. W. (1980) "The Semantic Interpretation of 
Nominal Compounds," Proceedings of the First Na- 
tional Conference on Artificial Intelligence, Stanford, 
August 18th-21st, 1980, 310-312. 

Fleischer, W. (1982) Wortbildung der deutschen Gegen- 
wartssprache, Niemeyer, Tfibingen. 

Isabelle, P. (1984) "Another Look at Nominal Compounds," Proceedings of the lOth International Conference on Computational Linguistics and the 22nd Annual Meeting of the Association for Computational Linguistics, Stanford, July 2nd-6th 1984, 509-516. 

Johnson, R. and M. Rosner (1989) "A Rich Environment 
for Experimentation with Unification Grammars," 
Proceedings of the Fourth Conference the European Chapter of the Association for Computational 
Linguistics, Manchester, April 10th-12th 1989, 182-189. 

Kaplan, R. M., K. Netter, J. Wedekind and A. Zaenen 
(1989) "Translation by Structural Correspondence," 
Proceedings of the Fourth Conference of the European Chapter of the Association for Computational 
Linguistics, Manchester, April 10th-12th 1989, 272-281. 

Russell, G., A. Ballim, D. Estival and S. Warwick-Armstrong (1991) "A Language for the Statement of 
Binary Relations over Feature Structures," Proceedings of the Fifth Conference of the European Chapter of the Association for Computational Linguistics, 
Berlin, April 9th-llth 1991, 287-292. 

Shieber, S. M. (1986) An Introduction to Unification- 
Based Approaches to Grammar, CSLI, Stanford. 

Sparck Jones, K. (1983) "So what about parsing com- 
pound nouns?," in Automatic Natural Language 
Processing, K. Sparck Jones and Y. A. Wilks, eds., 
Ellis Horwood, Chichester, 164-168. 
