CHAPTER 2. APPLICATION OF HOUSE’S MODEL FOR
TRANSLATION QUALITY ASSESSMENT
2.1. Presentation of the model
Julian House is a German linguist. She first proposed her model for translation quality assessment in 1977 in A Model for Translation Quality Assessment. The
original model attracted many criticisms which are said to have been tackled in her revised model in 1997 – Translation Quality Assessment: A Model Revisited.
2.1.1. An overview of the model
Not until the 1970s did the translation science begin to deal with translation criticism, which in House’s words is translation quality assessment. House was
among the theorists who tried to provide a scientifically based framework for translation quality assessment. For her, translation is “the replacement of a text in the
source language by a semantically and pragmatically equivalent text in the target language” 1997: 31. To be equivalent, the TT has to preserve the meaning of the
ST. There are three aspects of this meaning: a semantic aspect, a pragmatic aspect, and a textual aspect. House considers equivalence to be the fundamental criterion of
translation quality and argues that it is functional, pragmatic equivalence that should be achieved for ‘meaning’ to be preserved across two different languages. By
pointing out that ‘any two linguistic items in two different languages are multiply ambiguous’ and that ‘languages cut up reality in different ways’, House stresses that
functional, pragmatic equivalence “is the type of equivalence which is most appropriate for describing relations between original and translation”. But in the
model, poetic-aesthetic texts are excluded. As written in House’s report on Meta 2001: 247, the assessment model is
“based on Hallidayan systemic-functional theory, but also draws eclectically on
Prague school ideas, speech act theory, pragmatics, discourse analysis and corpus- based distinctions between spoken and written language”. And after being revised,
House’s model still aims at providing for ‘the analysis and comparison of an original and its translation on three different levels: the levels of LanguageText, Register
Field, Mode and Tenor, and Genre’ House, 2001: 247, or for the discovery of ‘mismatches’ between the TT and the ST on the same three levels.
2.1.2. Operation of the model
A key concept in the operation of House’s model is the context of situation or ‘situational dimensions’. The author states that the function of a text is “the
application or use which the text has in the particular context of situation” p.36. There are two components of a text’s function: the ideational and the interpersonal
components. And functions of texts are different from functions of language. A translation is functionally equivalent to its ST if it does not only match its ST in
function, but also employs equivalent situational-dimensional means to achieve that function.
The ‘situational dimensions’ can be collapsed into two sections: Dimensions of Language User and Dimensions of Language Use.
A. Dimensions of Language User 1. Geographical Origin
2. Social Class 3. Time
B. Dimensions of Language Use 1. Medium: simplecomplex
2. Participation: simplecomplex 3. Social Role Relationship
4. Social Attitude 5. Province
Each of these situational dimensions is realized through syntactic, lexical and textual means. Textual means refers to:
1. theme-dynamic: thematic structure and cohesion; 2. clausal linkage: additive and, in addition, adversative but,
however, alternative, causal, explanatory, illative relations, etc.; 3. iconic linkage: parallelism of structures.
From House’s assumption that in order to make qualitative statements about a TT, the TT’s textual profile must be compared with the ST’s textual profile, her
model for translation quality assessment operates as follows: 1. ST is analyzed in details to produce a profile of the ST register with text-
specific linguistic correlating to the situational dimensions syntactic, lexical, and textual means.
2. A description of the ST genre realized by the register is added. 3. A statement of function of the ST is made, including the ideational and
interpersonal component of that function. 4. A similar profile and statement of function is made of the TT.
5. The TT profile is compared to the ST profile using the schema below and a statement of ‘mismatches’ is produced, categorized according to genre
and to the situational dimensions of register and genre. These dimensional errors are distinguished from denotative mismatches or target system
errors. 6. A ‘statement of quality’ is made of the translation.
7. Finally, the translation is categorized into either overt translation or covert translation.
Munday, 2001: 93
The schema for analyzing and comparing original and translation texts House, 1997: 108:
Register in House’s model resembles Halliday’s register analysis but has some additions. Field refers to subject matter or topic, social action or social activity and
covers the specificity of lexical items. Tenor refers to the nature of the participants and the relationship between them, including the ‘text producer’s temporal,
geographical and social provenance as well as his intellectual, emotional or affective stance his “personal viewpoint” […] ‘social attitude’, i.e. different styles formal,
consultative and informal. Mode refers to both the channel – spoken or written, and the degree of participation between writer and reader House, 2001: 248.
But register analysis alone is not sufficient for a statement of text function to be made. The analysis of the text, may it be ST or TT, has to incorporate Genre –
subject matter and social
participant relationship + author’s provenance
and stance + social role relationship
+ social attitude
+ medium simplecomplex
+ participation simplecomplex
Individual textual function
which is ‘the conventional text type’ associated with ‘a specific communicative function’. Genre is conditioned by the socio-cultural environment and itself
determines other elements in the systemic framework. As House says, Genre ‘connects texts with the “macro-context” of the linguistic and cultural community in
which texts are embedded” 2001: 248. There are criticisms that House’s analytical schema is over-complicated. There
are even more criticisms regarding House’s distinction of two types of translation: overt translation and covert translation. An overt translation is “one in which the
addressees of the translation text are quite ‘overtly’ not being directly addressed” 1997: 66. A covert translation is “a translation which enjoys the status of an original
source text in the target culture” 1997:69. In overt translation, equivalence is necessary at the level of LanguageText and Register, as well as Genre. In covert
translation, equivalence is desirable at the level of Genre and the individual text function. But in covert translation, the translator has to apply what is called a
“cultural filter” to modify cultural elements and to give receivers the impression that the TT is an original. The translator assessor also has to make use of this ‘cultural
filter’ in judging the quality of a covert translation.
2.2. Application of the model