Summary
Frame analysis is a relatively new methodological approach which shows how people understand 
activities or situations. It originated in sociology and its application to translation has not been 
considered practically or theoretically yet, though the advantages may be manifold. Consisting 
of two main parts, this paper presents a methodology for frame identification and analysis, and 
suggests this be applied to the translation of pragmatic texts. $e first part presents the concepts 
of frame and frame analysis as they appear in literature and as they are interpreted in this paper 
for translation purposes. $e second part focuses on the exemplification of a methodological 
framework which includes the integration of frames into the translation process. It is shown 
that by using frames, translators can obtain the cognitive image of the text, create various 
versions of the source text in the target language, and use translation strategies consistently and 
transparently. 
Key words: frame analysis, information universe, translation process, translation 
methodology 
 Povzetek
Okvirna analiza kot razmeroma nov metodološki pristop pojasnjuje, kako ljudje umevajo 
dejavnosti in razmere. Izvira iz sociologije, njena vloga pri prevajanju pa doslej še ni bila praktično 
ali teoretično obdelana kljub njenim mnogoterim prednostim. Članek v prvem delu predstavi 
koncepte okvira in okvirne analize v luči obstoječe literature in jih nato aplicira na potrebe 
prevajanja. V drugem delu pa s primeri podkrepi metodološki pristop v smislu vključevanja 
okvirov v proces prevajanja.  Raziskava dokaže, da si lahko prevajalci s pomočjo okvirov ustvarijo 
kognitivno sliko besedila, predstavljajo različne oblike izhodiščnega besedila v ciljnem jeziku ter 
dosledno in nazorno uporabljajo prevajalske strategije.
Ključne besede: okvirna analiza, univerzalna informacija, proces prevajanja, prevodna 
metodologija
 DOI: 10.4312/elope.7.1.121-130

According to recent claims and desiderata in translation studies, the more translators know about 
the structure and the dynamics of discourse, the more readily and accurately they can translate 
both the content and the spirit of a text (Nida 1997, 42). Similarly, international research projects 
highlight directions of research which aim at helping translators make reasonable and consistent 
decisions as to the relevance and reliability of source text features in the target text (Gerzymisch-
Arbogast 2005, 7). In line with such voices and based on a deficit analysis whose results were 
published recently (Dejica 2009), I have created a model for source text analysis (Dejica 2006, 
103−10) which analyses the Information Universe (IU) constituents, i.e., information carriers 
in texts, from various perspectives: atomistic, hol-atomistic, and holistic (Gerzymisch-Arbogast 
2006). Its main aim is to facilitate text understanding and, implicitly, translation. 
In my interpretation of the text perspectives (Dejica 2008, 77−91), the atomistic perspective 
implies identification and analysis of salient individual IU constituents in texts, whereas the 
hol-atomistic perspective implies the analysis of all possible relations at text level between IU 
constituents, such as cognitive, semantic, lexico-grammatical and syntactic. Finally, the holistic 
analysis implies identification, establishment and analysis of various possible cultural or generic 
relations above text level, which are established between such constituents and other relevant 
constituents from different IUs.   
$e current paper deals with the presentation of one type of analysis suggested to be performed from 
a hol-atomistic perspective during the translation process, i.e. the analysis of cognitive relations, 
which can be represented in texts in the form of information frames. In the first part, the concepts 
of frame and frame analysis are presented as they appear in the literature and as they are seen in this 
paper. In the second part, a suggested methodology for frame identification, analysis and application 
to translation is illustrated. $e methodology is explained and exemplified on a pragmatic, i.e., non-
literary, text. $e last part resumes the consideration of the advantages of using the model. 
 Very broadly, the term “cognition” is used to define the act of knowing, or knowledge. Sociologists 
and psychologists have shown that people organize and store their knowledge of the world as fixed 
data structures, one such form being the frame (Goffman 1974, Charniak 1976, König 2007). 
$e concept of frame is basically synonymous to Goffman (1974), who was the pioneer in frame 
identification and analysis from a sociological perspective. $e excerpt below illustrates Goffman’s 
(1974) idea of frame as it appears in Frame Analysis – An Essay on the Organization of Experience, 
his well-acclaimed and, at the same time, very controversial book (Lemert 1997, xii): 

It has been argued that a strip of activity will be perceived by its participants in terms of the rules or 
premises of a primary framework, whether social or natural, and that activity so perceived provides 
the model for two basic kinds of transformation - keying and fabrication. It has also been argued 
that these frameworks are not merely a matter of mind but correspond in some sense to the way in 
which an aspect of the activity itself is organized - especially activity directly involving social agents. 
Organizational premises are involved, and these are something cognition somehow arrives at, not 
something cognition creates or generates. Given their understanding of what it is that is going 
on, individuals fit their actions to this understanding and ordinarily find that the ongoing world 
supports this fitting. $ese organizational premises - sustained both in the mind and in activity - I 
call the frame of the activity. (Goffman 1974, 247)
In the same vein, Charniak (1976) defines frame as 
A static data structure about one stereotyped topic, such as shopping at the supermarket, taking 
a bath, or piggy banks. Each frame is primarily made up of many statements about the frame 
topic, called “frame statements” (henceforth abbreviated to FS). $ese statements are expressed in 
a suitable semantic representation, although I will simply express them in ordinary English in this 
paper. (Charniak 1976, 42)
In more basic terms and more recently, König (2007) provides his own interpretation of frames 
based on Goffman’s definition. $us, according to him, 
Frames are basic cognitive structures which guide the perception and representation of reality. On 
the whole, frames are not consciously manufactured but are unconsciously adopted in the course 
of communicative processes. On a very banal level, frames structure which parts of reality become 
noticed. (König 2007)
König’s definition is accompanied by an exemplification of a frame in a real life situation, quoted 
below: 
For example, a group of persons lined up in an orderly fashion at the side of a road might evoke 
the frame “bus queue” in a passer-by. $is particular frame structures perception in the way that 
attention is paid to the orderly arrangement of people in a line, which is one indicator of the “bus 
queue frame” and might have actually triggered it. $e frame also directs attention to other latent 
frame elements, such as a bus stop sign. At the same time, it deflects attention from clothing style, 
body shape, or communications among the presumed prospective bus passengers. (König 2007)
T o put it broadly, from these definitions and exemplifications, frames can be considered as data 
or information structures which gather typical individuals, actions and activities, i.e. constituents 
in general, in a situation. Basically, examples of frames could be limitless, e.g., university-frame, 
car-frame, computer frame, etc.  
$ese frames are static representations showing what constituents are associated, i.e., a university 
frame may consist of students, lecturers, classes, learning material, etc. As for how these constituents 
are associated, frames are seen as being made up as recognizable structures of relevancies. What 
and how constituents are associated in a frame is accounted for cognitively, based on contextual 
factors and on each individual’s knowledge background.
In the last decades, frame analysis has evolved as a discipline in its own rights, being mostly used 

in communication studies. $e evolution of frame analysis is explained by Lemert (1997), who 
admits that at first, 
Sociologists never quite knew what to make of Frame Analysis (1974) and other later works implicated 
in Goffman’s own linguistic turn, while it is precisely Frame Analysis and Forms of T alk (1981) that 
drew the most attention outside the field. (Lemert 1997, XV)
T o my knowledge, its application to translation and its integration into translation studies, ‘outside 
the field’ as Lemert (1997) puts it, have not been yet considered practically or methodologically. 
It is my belief that both translators and translation scholars could benefit from the exploration of 
this inter-disciplinary, yet virgin ground. 
Since frame identification is usually based on intuitive grounds and on general background 
knowledge, and since recognizable structures of relevancies could be accounted for differently by 
different translators, I consider that in relation to translation, frame identification and its subsequent 
analysis should be reconsidered so that translators may approach them objectively and consensually. 
$e suggested approach to frame identification and analysis presented here combines pragmatic 
identification of information and hol-atomistic analysis (Dejica 2008, 77−91). It is meant to offer 
translators a clear cognitive image of the text in the form of an information frame, which can be 
used to serve various translation purposes. $e suggested approach consists of a series of steps which 
are integrated into a translation process and exemplified in Part 3 of this article. 
In relation to translation, I consider that every text displays two different types of frames: a 
generic frame, conceptual in nature, and an information frame, which is the materialization in 
the text of the conceptual frame.  
Every translator, if more or less experienced, has what I call his/her own generic image of the text to 
be translated. $e generic image consists of a series of genre specificities (Bhatia 1993, Swales 1993), 
i.e., textual structure and organization, thematic progression, word order, modes, content, etc. $ese 
generic or genre specificities are not seen or made explicit by the writer, but known by the translator 
from his/her experience or training. For example, in the case of a project proposal, the generic 
frame is formed of the project aims, historical development of the organization, sustainability of 
the project, budget, etc. $e budget section is richer in figures than the development section (which 
means less or no text to be translated), and so on. I call this image the generic frame of the text. It is 
conceptual in nature and setting it up and visualizing it takes place before the actual reading of the 
text, usually from the background information the translation might (or should) have on particular 
genres, or from preliminary discussions with the client. 
$e information frame or information universe frame is in my conception a materialization in 
the text of the generic frame; it is a static representation which presents a situation, process, 
product, etc., and which consists of various Information Universe constituents (Dejica 2006, 
103−10), which are as different as are texts. Hence the multitude of information frames which 
can be identified and analysed for each different text. If usually the generic frame of a text is 
only one for each genre (it may display variations in the case of sub-genres), its materialization 
is different in different texts. In other words, the project proposal as a genre usually has the 

same structure, degree of specificity, etc., but different informational content, i.e. it may refer 
to ways of improving the environment, saving energy, preserving nature, etc., materialized in an 
information frame specific for each proposal.
 T ranslation is seen here as “an activity which transfers into a target text – with a specific 
purpose in mind – the writer’s intention expressed in a source text” (Dejica 2009a, 131). I use 
‘transfer’ with a double connotation: that found in Shuttleworth and Cowie (1997) and Hatim 
and Munday (2004) to imply that I see translation as process, and that found in Nida and 
Taber (1969) and Gerzymisch-Arbogast (2005) to refer to the second phase of the translation 
process, i.e., that of transfer, where the ‘material’ analysed in the reception phase is transferred 
into the mind of the translator and compared for translation purposes. Reception, transfer and 
reproduction are the three phases of translation on which the following suggested methodology 
for approaching cognitive relations, and implicitly information frames, is based. $e approach 
to frame identification and analysis for the translation of pragmatic texts suggested here is thus a 
multi-phase process which combines pragmatic identification of information and hol-atomistic 
analysis, and which consists of a series of phases and steps. For exemplification and analysis, I 
will use the Ptolemy Project text (below), a sample of a pragmatic text taken from an extended 
project description: 
Project Ptolemy Objectives
$e project aims to develop techniques supporting heterogeneous modelling, including 
both formal “meta-models” and a software laboratory for experimenting with heterogeneous 
modelling. In this context, it will explore methods based on dataflow and process networks, 
discrete-event systems, synchronous/reactive languages, finite-state machines, and 
communicating sequential processes. It will make contributions ranging from fundamental 
semantics to synthesis of embedded software and custom hardware. 
 (Ptolemy Project, http://ptolemy.berkeley.edu/objectives.htm, May 25, 2006)
• Phase 1. Reception. $e first translation phase, synonymous to text understanding, 
implies the following sequence of steps:
Step 1. Establishment of the generic frame of the text
As explained before, the generic frame is conceptual in nature and consists of all the knowledge 
a translator has on a particular genre. $e specialized educational background and the practical 
experience of the translator are essential in establishing this conceptual frame. In this respect, 
familiarity with results of empirical studies on genre specificities can definitely be a plus, even 
though such studies are relatively few and their results are not standardized. $e application 
of genre analysis to translation is also a relatively new inter-disciplinary approach, which is 
recommended by many voices (T rosborg 2000).  
Step 2. Identification of the Information Universe constituents. 
All the constituents expressed in the source text form what I call the Information Universe of the 

text (Dejica 2006, 103−10). I use the term ‘universe’ as in the natural sciences, where it stands 
for the sum of everything that exists in the cosmos. Just as in science, in this approach, universe 
stands for the sum of all the information that exists in a text. $e IU constituents are carriers of 
information which structurally can be divided into a two-part information system, which in my 
approach is formed of $emes and Rhemes. 
For the identification of $emes and Rhemes, I suggest the use of a pragmatic $eme-Rheme 
(PTR) model. As the name implies, the model uses pragmatic parameters for the identification 
of information constituents, i.e. $emes and Rhemes, such as shared background knowledge 
of the sender and receiver, scope of attention, etc. $e model and its application are described 
in detail in a series of publications (Dejica 2006, 2008, 2009a) and resuming them here would 
exceed the aim and length of the paper. For the sake of the subsequent presentation, I present 
here only the results of the analysis, in the form of identified $emes and Rhemes: 
1. $eme – given information: the project 
2. Rhemes – new information: (i.e. the objectives proper): development of techniques supporting 
heterogeneous modelling, including both formal “meta-models” and a software laboratory for 
experimenting with heterogeneous modelling; exploration of methods based on dataflow and 
process networks, discrete-event systems, synchronous/reactive languages, finite-state machines, 
and communicating sequential processes; making contributions ranging from fundamental 
semantics to synthesis of embedded software and custom hardware.
Step 3. Hol-atomistic analysis of the Information Universe constituents. 
By visualizing the identified IU constituents, the translator draws hol-atomistic cognitive relations 
between the rhematic information constituents and materializes them in the form of a frame, 
which in the selected example is a computer frame. $e computer frame is not named explicitly 
in the text but is deduced from the sum of all the rhematic information in the IU: heterogeneous 
modelling, dataflow, process networks, discrete-event systems, synchronous/reactive languages, finite-
state machines, communicating sequential processes, embedded software, and custom hardware. 
Step 4. Narrowing the Hol-atomistic analysis.  
$is step could be performed as part of Step 3, but for reasons of clarity I treat it separately; 
during this step, the translator basically narrows the analysis by drawing structural cognitive 
relations between the elements of the previously identified IU frame (IUF) with the aim of 
finding possible secondary meaning categories within the frame. Such possible secondary 
meaning categories, subordinated to a main IU frame, can be classified into secondary frames or 
sub-frames (SF). In the Ptolemy Project example, there are two such SFs, i.e., the hardware and 
software components of the computer frame: hardware frame (networks, systems, machines) and 
software frame (modelling, dataflow, languages, processes, software).
Step 5. Resuming the Hol-atomistic analysis.  
Once the main IUF and its possible SFs are identified, the translator resumes the process (Steps 
3 and 4) so as to identify other possible IUFs and SFs. $is is a very useful step since smaller 

frames, usually not identified at first sight, may be discovered and established as well. Usually 
short texts can display two or three information frames. In the analysed example, one more IUF 
can be identified, i.e. a research-frame, from the sum of the processes used to put the computer-
frame into practice: techniques, modelling, laboratory, experimenting, and exploration.
Step 6. Visualizing the identified information frames.  
In this last step of Phase 1, once all the possible IUFs and SFs have been identified (Steps 3-5), 
the translator draws superordinate and subordinate relations between them and completes thus the 
cognitive image of the text to be translated. In the Project Ptolemy example, the Information Universe 
frames together with the sub-frames can be represented as a network of relations as follows:
 
 • Phase 2. Transfer. In the second translation phase, based on the previous analysis, the 
translator decides upon a final Information Universe frame which will be used for creating 
the target text in Phase 3. 
Step 7. Establishing the Information Universe frame in the target text.
$is is a decisive step in the translation process. When establishing the IU frame to be used in the 
creation of the target text, the translator must take into account the translation situation which 
may consist of a series of factors, such as the writer’s intention, the translation purpose, the target 
audience, the target language peculiarities, and the client’s requirement. Sometimes, some of 
these factors can be established prior to starting the process upon preliminary discussions with 
the client. If the client does not have any specific requirements, based on the other factors, the 
translator can now decide whether the target text will be source- or target-language oriented, 
i.e., whether it will preserve the flavour of the original or whether it will be adapted to the target 
language specificities. 
Step 8. Building the final frame using translation strategies.

Once the translator has reached a decision and the translation situation is clear, s/he builds the 
final frame using the most adequate translation strategies (Leppihalme 1997, Newmark 1988, 
Delisle 1999, Chesterman 2008). For exemplification purposes and assuming that the target 
language is Romanian, based on the source-text frame visualized in Step 6, the following two 
possible target-text frames can be obtained to suit two different translation situations: 
a. T ranslation situation 1: source-language oriented text
 
Visualisation of structural cognitive relations between the information universe constituents in the target text (source-langua
 b. T ranslation situation 2: target-language oriented text
 


In the two translation situations presented above, different translation strategies were used to 
render the same frame constituents from English into Romanian. In the first situation, to create 
a source-language oriented text, one which preserves the flavour of the original, borrowing, loan 
(Vinay and Darbelnet 1995) or foreignization (Venuti 1998) were the predominant strategies 
used; in the second one, to create a target-language oriented text, practically a text where the frame 
constituents would be identical to those normally used in the target language by a native speaker/
writer, domestication (Venuti 1998) or calque (Vinay and Darbelnet 1995) were considered to 
be the most appropriate translation strategies. In Figs. 2 and 3 above, the application of these 
strategies is highlighted with bold characters. 
• Phase 3. Reproduction. In the last phase of the translation process the target text is created. 
Step 9. T ranslation 
$e translator translates now the text using the final frame established in Step 8. $is is relatively 
a rather easy step, but one which should be used in conjunction with other analyses of the source 
text, i.e., atomistic, hol-atomistic and holistic (Dejica 2008, 77−91). 
 Applying information frames to translation is not sufficient as a key to felicitous translation. 
$eir application should be used in conjunction with the other suggested analyses, i.e. semantic, 
lexico-grammatical, syntactic, cultural and generic, meant to be performed from the three 
different atomistic, hol-atomistic and holisitc perspectives. 
Some more words of caution are in order. In practice, due to this rather complex methodological 
approach, identifying, analysing and re-creating information frames may be sometimes considered 
to slow down the translation process and eventually may prove not to be cost-effective. Moreover, 
their application would not be suitable for extensive texts, for the very same reasons. 
However, information frames may prove extremely helpful in the case of short problematic texts, 
highly specialised texts, or those rich in terminological content. In this case, their contribution 
to translation is undeniable: applied from a hol-atomistic perspective, information frames are 
extremely helpful for visualizing cognitive relations in texts, which ultimately leads to text 
understanding. Other major advantages of using information frames for the translation of 
pragmatic texts include but are not limited to the possibility of (1) creating source- or target-
oriented versions of the original text, all different, yet all ‘correct’ depending on the translation 
situation, (2) using translation strategies consistently and transparently throughout the entire text, 
and (3) applying them to any other language pairs, either for professional or didactic purposes. 

Bhatia, V.J. 1993.  London: Longman. 
Charniak, E. 1976. Organization and inference in a frame-like system of common sense knowledge. In , eds. R. Schanck and B. Nash-Weber, 42−51. Cambridge, 
Massachusetts. 
Chesterman, A. 2008.  Available at  (accessed 15 November 2008).
Dejica, D. 2006. Towards a New Model for the Identification of Thematic Information in Discourse. In , ed. R. Superceanu and D. Dejica, 
103−10. Timişoara: Editura Politehnica.
−−−. 2008. Information Universe and Text Perspectives in Translation. In  Proceedings of the International Conference Tras.Re.P ., ed. R. Croitoru, issue 3/2009: 77−91. Galaţi: Galaţi University 
Press.
−−−. 2009a. Identifying and Analysing Theme-Rheme Relations for Discourse Production and Translation.  6: 128−36. Timisoara: Mirton.
−−−. 2009b. Probing for Relevance: Information Theories and Translation.  V(1−2): 81–99. 
Delisle, J., ed. 1999.  Amsterdam: John Benjamins.
Gerzymisch-Arbogast, H. 2005. Introducing Multidimensional Translation. In  Proceedings 
of the Marie Curie Euroconferences − Saarbrücken 2−6 May 2005, eds. H. Gerzymisch-Arbogast, and S. Nauert. Available at  (accessed 1 March 2008).
−−−. 2006. Text Perspectives and Translation. Available at  (accessed 1 March 2009).
Goffman, E. 1986.  Boston: Northeastern University Press. 
Hatim, B., and J. Munday. 2004.  London & New York: Routledge.
König, T. 2007 . Frame Analysis: Theoretical Preliminaries. In  Available at 
 (accessed 15 March 2009).
Lemert, C. 1997 . Goffman. In , ed. C. Lemert and A. Branaman, ix – xiii. Massachussets: Blackwell Publishers Ltd.
Leppihalme, R. 1997 .  Clevedon: Multilingual Matters. 
Minsky, M., 1976. Minsky’s frame system theory. In , ed. R. Schanck and B. Nash-Weber, 104−16. Cambridge, Massachusetts.   
Newmark, P . 1988.  New York: Prentice Hall. 
Nida, E.A. 1997 . The Principles of Discourse Structure and Content in Relation to Translating. In , ed. K. 
Klaudy and J. Kohn, 37−42. Budapest: Scholastica. 
– – –,  and C.R. Taber. 1969.  Leiden: E.J. Brill.
Shuttleworth, M., and M. Cowie. 1997 . Manchester: St. Jerome.
Swales, J. 1993.  Cambridge: Cambridge University Press.
Trosborg, A. 2000. Text and Genre Analysis for Translators: Structure, Content and Function of the Model.  7(3): 229−44. 
Venuti, L. 1998.  New York: Routledge.
Vinay, J.P ., and J. Darbelnet. 1995.  Trans. J.C. Sager 
and M.J. Hamel. Amsterdam / Philadelphia: John Benjamins.