4.

The priority of purpose (skopos theory)

4.1 Introductory remarks

An action aims to achieve a goal and thus to alter the current state of affairs.
The motivation for such an action is that the intended goal is estimated to be
of greater importance than the current state of affairs. Sometimes, an action
is preceded by a chain of motivations: if somebody acts under compulsion,
this aim may be estimated to be of lesser value than the current state of affairs
but compliance may be less damaging than resistance. An action is always
preceded by (conscious or unconscious) expectations about a future situation
in comparison to how the current situation has been assessed. By summarizing
all the requirements for an action, we are presupposing that it makes sense
for an agent to choose one aim from all the possible culture-specific options
in a particular situation. Incidentally, a general theory of action need not be
explained in detail; see the existing literature on this topic, above all Rehbein
(1977) and Harras (1978). With regard to “action”, cf. Harras (1978: 19);
with regard to action requirements, cf. Harras (1978: 28-31); with regard to
“intention”, cf. Biessner (1982).
We shall now examine the difference between a general theory of action
and a general theory of translational action (as a subcategory of the former).
A theory of action begins with a specific situation which is assessed by a
particular person in a particular way. This person then acts in such a way that
his action can be justified by his assessment of the situation (we shall not go
into the details). A theory of translational action begins with a situation that
always includes a preceding action, i.e. the source text; here, the question is
not whether and how somebody acts but whether, how and in what respect
the previous action is continued (translated/interpreted). Seen in this light, a
theory of translational action is a complex theory of action.
Consequently, translational decisions are based on a fundamental rule
which not only determines whether something is transferred and what is trans-
ferred, but also how it is transferred, i.e. according to which strategy.

A translational action is governed by its purpose.

Each utterance is an expression of the cognitive state and the intentions of

the speaker.42
Each discourse is more or less clearly directed at a goal (intentional)
and, as such, it is an instrument for pursuing intentions.43
42
Jede Äußerung ist Ausdruck des jeweiligen Bewußtseinszustandes und der Absichten
eines Sprechers. (Stern 1974: 67)
43
Jede Rede ist mehr oder minder zielgerichtet (intentional), als solche ist sie ein Instrument
zur Verfolgung von Absichten. (ibid.: 75)
Katharina Reiß and Hans J. Vermeer 87

changed over the course of time (sometimes a reason for ‘intergenerational

conflict’). With regard to the acceptance of norms, societies are not one uniform
group, i.e. they consist of discrete subgroups.
To explain the acceptance (or not) of norms, we shall use the analogy of
games and their rules. As a member of a society, each person must play and
they can choose a game from among a number of possible games. The rules
of the chosen game must be observed as long as the game lasts. The players
determine the duration of the game. A chosen game can be abandoned in
favour of another one.
Acting can be described as re-acting (in the broadest sense) to a given
situation. Actions are regarded as recurrent when common features can be
observed in situational factors. Norms are rules for recurrent behaviour (act-
ing) in specific types of situations. Norms are culture-specific.
Therefore, the type of (assumed) situation is a culture-specific constant in
acting. In a particular situation, acting is determined by an assessment of the
factors of the situation at hand. Acting is regarded as reasonable if a claim can
be made that it is appropriate, in line with the norms of the given culture (!)
and the situation, by the person who is in a position to make this claim.
We shall assume that every person wants to act reasonably (i.e. based on
reasonable motivating factors) in a given situation, so that they can claim
that their behaviour is appropriate for the situation (these claims need not be
expressed, or expressible, in well-formed sentences). Acting is ‘intentional’ or
‘purposeful’ in two ways: it is intended to be appropriate to the situation and
it is intended to achieve an aim in a given situation, e.g. to adjust oneself to a
situation or to change a situation to mould it to one’s own purpose.
Acting may therefore be described as a function of two factors: the assess-
ment of a given situation and the intention (or function/purpose of the action)
deriving from it:

A = f (sit, int)

Norms are established at the level of type and not of token. In individual
situations, there is usually a choice between a limited number of options (in
exceptional cases, the number of options may be 1). There is no strict or pre-
dictable sequence of individual utterances for a particular situation, but there
is a framework for the sequence of utterance types. Kallmeyer and Schütze
(1976: 2) confuse type and token when they claim that there is a fixed sequence
of utterances in individual situations. They give the following example:

If one person begins a telephone call with the question What are you
doing at the moment?, the other person must respond in order to avoid
a ‘severe crisis’. A person who were to answer with another question,
e.g. Why do you want to know?, would be asking for trouble in this
situation (cf. Kallmeyer and Schütze 1976: 24, note 4).
88 The priority of purpose (skopos theory)

We do not agree with Kallmeyer and Schütze because we believe that the rules
regulating interaction are not so strict. ( 3.)
Norms stipulate that people should act and how they should act. But they
leave a certain margin for choice. What is crucial is that the reaction to a situa-
tion be such that it can be interpreted as reasonable (of course, each participant
of an interaction may be asked for an independent interpretation, but this is
something we shall not go into at this point; cf. Schnelle 1973: 37). How a norm
is complied with is less important than the fact that the interaction partners try
to comply with it. It is the function of the action which is crucial.
Eykman (1977: 35-42) shows that images or words can be replaced by
other images or words without actually changing the text function. Eykman
speaks of “modifications” (Abwandlungen) instead of variants. For transla-
tion, this means that (1) under certain conditions, modifications are legitimate,
and (2) the conditions are defined by culture-specific norms, e.g. that in order
to be considered adequate, modifications should have the same degree of
conventionality as the modified image or word.

What we do is less important than the purpose of the action and its
being achieved.

An action is successful if it can be interpreted as adequate or reasonable with

regard to the situation. As we have already stated, this interpretation is first
required of the agent (producer) himself: he must state what his ‘intention’ was.
We have pointed out that not every intention can be (optimally) transformed
into an action (you may hit your finger with the hammer before you actually
hit the nail). On the other hand, the interaction partner (the recipient) may also
try to find an interpretation for the producer’s action. The recipient’s interpre-
tation may differ from that of the producer. Each one tries to anticipate the
explanation of the other and will take it into account in his action (‘reflexive
co-orientation’). (With regard to the trans-individuality of interpretations cf.
Schnelle 1973: 37.) Whether there has been a ‘success’ is therefore declared
independently by the producer and the recipient and remains separate for both
(and perhaps even for a third party).
Every action can be assigned a value which for the producer depends on
its intention (intended function) and for the recipient on its interpretation
(interpreted function).
What a theory of interaction primarily needs is not so much a two-value
logic theory of truth values as a many-valued (fuzzy) logic theory. It is in the
framework of such a many-valued logic theory that we must consider the ‘truth
value’ of an action, as the ‘value’ of an action in a given situation depends,
among other things, on the relationship between the action and reality (as a
phenomenon; cf., for example, an intended/recognized lie or a mistake. With
regard to such a truth value theory cf. Marten 1972, among others.)
Katharina Reiß and Hans J. Vermeer 89

An action can be considered (completely) successful (for both parties) if

the values assigned to it by the sender and the recipient are within the permis-
sible value parameters set for each case so that neither of the two ‘protests’.
Based on what they agree to consider an adequate action, both parties will
have to negotiate with each other and set the degree of difference which will
be regarded as significant in a given situation. When evaluating an action,
the recipient must, as we have already noted, take the producer’s intention as
an intention-in-the-given-situation-for-the-producer into account (cf. Grice
1957; Alston 1964; Searle 1967; and Marten 1972: 174, note 352, who agrees
with all of them.)
In other words, each action, and its interpretation, presupposes a ‘theory’
about the action and its conditions. An action can be regarded as successful
if the producer’s theory about the action and the recipient’s theory about the
interpretation do not differ significantly.
With regard to the manner of acting, we should make a distinction between
(1) how the carrying out of an intention is planned, i.e. what is done in the
planning phase (which ‘theme’ is chosen in order to achieve an intention); and
(2) how the theme is put into practice, i.e. what is done in the implementation
phase (when the ‘rheme’ is carried out).

For example: let us assume that the skopos of our action is to distract
someone else from a pressing anxiety. To achieve this skopos, we
choose to distract the other’s attention with the ‘the­me’ of an interesting
phenomenon, a concert. In order to achieve this intention, we suggest
attending the concert.

4.3 Summary

In the previous section, we attempted to look at the issues discussed in chap-

ters 3. and 4. from a different angle. To sum up, we can say that the skopos of
an action takes precedence over the mode of action, i.e. the purpose determines
whether, how and what is done.

Given that translational action is a specific form of interaction, it is

more important that a particular translational purpose be achieved
than that the translation process be carried out in a particular way.

For example: let us assume that the skopos is ‘Translate Genesis 1 as

a magical text’ (cf. Buber and Rosenzweig 1954;  3.1., Example 1).
In this case, it is more important to reproduce the sound of the words
as closely as possible than to render their meanings. If the skopos
is ‘Translate Genesis 1 as an aesthetic text’, it is more important to
achieve the aesthetics of the text in line with the standards of the target
90 The priority of purpose (skopos theory)

culture (!) than to preserve the wording of the source text. If the skopos
is Translate the Bible as an informative text, it is important to convey
the meaning of the text (as far as this is possible), either for theologians
or for a general audience, etc. (cf. Kassühlke 1983). There is no such
thing as ‘the’ translation of a text; the results of the translation process
will vary according to their skopoi.

4.4 The skopos rule

The highest rule of a theory of translational action is the ‘skopos rule’: any action
is determined by its purpose, i.e. it is a function of its purpose or skopos.

Trl. = f (sk)

In other words: for translational action, we can say that ‘the end
justifies the means’. There may be a number of elements in a set of
purposes (N > 1), in hierarchical order. Purposes must be justifiable
(reasonable).

4.5 The sociological rule

The intended audience (‘addressees’) or recipient may be described as a spe-

cific kind or subset of skopos. How an interaction is carried out depends, among
other things, on the relationship between the parties to an interaction.

For example: different forms of address are used for a professor or a

fellow student; the way a man is addressed in his role as professor will
differ from the way he is addressed in his role as father.

Consequently, we can arrive at the following sub-rule for the skopos rule:
the skopos can be described as a variable of the intended recipient (socio-
logical rule):

Sk = f (R)

In the general theory of translational action that we are aiming for, specific
limitations to the above rules are not taken into consideration, e.g. translat-
ing under duress.
As Reiß points out, it is “not only the purpose of a particular translation
which plays a role but also commissioners or publishers who may have a say
Katharina Reiß and Hans J. Vermeer 91

in a translation”44 (the influence printers had on the layout of a text in the

16th century is well known.)
In such cases, the general theory will either permit the translator to agree
to a skopos set by another person, or replace the translator’s own skopos by
someone else’s, which would require applying the skopos rule twice. In either
case, the skopos rule itself is not affected. Multiple (recursive) applications
of a rule may be described as successive or as simultaneous, with either an
amplifying or a dampening effect.

4.6 Phases in decision-making

Hella Kirchhoff (in a personal communication dated 6 April 1981) suggests

that the different methodological phases within the process of decision-
making should be distinguished from one another in order to facilitate their
application in translation practice.

(1) Setting the skopos

A skopos cannot be set unless the target audience can be assessed. If the target
audience is not known, it is impossible to decide whether or not a particular
function makes sense for them. That any ‘familiarity’ with the target audi-
ence can only be relative has been mentioned above ( 3.8.).

(2) Redefining the relevance of certain aspects of the source text according
to the skopos set
Whether such a redefinition is carried out before, during or after the process
of translational action may be subject to practical considerations.

For example: certain parts of a cultural history written in Spanish by a

Latin American author are rewritten if the translation addresses a wider
Central European audience with a different background knowledge.45
The adaptations could be made in the source text by a team of special-
ists before the translation is done, they could be made by a translator
who is also a specialist in the discipline during the translation process
or they could be made by a specialist after the translation is finished
(with regard to necessary linguistic modifications, cf. Reiß 1972).

44
[…] nicht nur der Zweck einer jeweiligen Übersetzung spielt [eine Rolle], sondern auch
Auftraggeber bzw. Verlage pflegen ein gewichtiges Wort mitzusprechen. (Reiß 1980a: 36-37)
45
Katharina Reiß herself translated a book dealing with Latin American history from Span-
ish into German: América Latina II. De la independencia a la segunda Guerra mundial, by
Gustavo Beihaut, reprinted as vol. 23 of the Historia Universal Siglo XXI (1986), a project
initiated by the German publisher Fischer Verlag. Reiß’s translation was published under
the title Süd- und Mittelamerika II. Von der Unabhängigkeit bis zur Krise der Gegenwart
as vol. 23 of the Fischer Weltgeschichte (Frankfurt 1965). (Translator’s note)
92 The priority of purpose (skopos theory)

Travel guides originally written for a non-German audience may have

to be rewritten, at least in part, for German readers because they are
interested in different aspects of a country (information provided by
Helga von Tobel).

(3) Accomplishing the skopos

The source text must be transferred functionally, taking the expectations of
the target audience into account.
Phases (1) and (2) require a familiarity with the target culture, whereas
phase (3) requires an additional competence in the target language.

4.7 Skopos hierarchies

Different sections of a source text may be translated for different purposes.

There may be a hierarchy of skopoi for texts and text sections (cf. Reiß
1976a).

4.8 Source-text skopos vs. target-text skopos

As we have pointed out several times, the skopos of the translatum may
be different from that of the source text. This can be justified by three
arguments:

(1) Translating/interpreting is an action which differs fundamentally from

producing a source text. Consequently, a translational action may serve
different purposes. It must be emphasized that the preservation of the source-
text purpose, which is often taken to be a defining feature of translation, is
a culture-specific rule and not a basic requirement of a general theory of
translational action.
In a large number of cases, it might even be impossible to fulfil this require-
ment, for example in a literary translation. Unless we want to limit the function
of a novel to the commonplace ‘entertainment’ function, we must assume that
the reasons Cervantes’ contemporaries had for reading Don Quixote were
not the same as those of Spanish-speaking readers today, and an American
reader’s reception of the story of the Seven Brothers (Finnish title: Seitsemän
veljestä) by the famous Finnish author Aleksis Kivi will be entirely different
from that of a Finnish reader.

(2) We have defined translational action as a kind of information offer

( 3.), i.e. information which is offered under the condition that the sender
expects it to be of interest (to contain something ‘new’) for the recipient.
This ‘novelty’ may consist precisely in the different skopos of the offer.
Katharina Reiß and Hans J. Vermeer 93

(3) Hirsch (1967, particularly p. 161) claims that the interpretation of a

text should identify exactly (and solely) what the implications encompassed
by “the purpose and the intention of the work” are, and that a text cannot
be fully understood unless these implications have been understood. No
more and no less. Even if we left aside the findings of reception history and
aesthetics for a moment and accepted the validity of this claim, we would
still have to state that translational action is a cultural and linguistic transfer.
Cultures and languages constitute independent systems in which the value
of each element is defined by its relationship with the other elements of the
same system (a system où tout se tient, where all elements are connected to
each other). In other words, cultures and languages are individual entities
and, therefore, texts, as systems consisting of parts of individual cultural
and linguistic systems, are individual entities as well. It is obvious that the
value of an element of one system that is transferred into another system is
bound to change because it is now related to the elements of the new system.
It is therefore impossible for the implications contained in the source text
to appear in exactly the same form in the target text (Koschmieder 1965,
Heger 1971 and Söll 1971 claim that it is possible at least for exclusively
cognitive features, but this is a purely methodological claim). Thus, if the
result is bound to be a ‘different’ text, we can only demand that it be as close
as possible to the source text. This may even be achieved precisely through
a change of function.

For example: for Homer’s contemporaries, the Iliad was what televi-
sion soaps are for the general public today. They could identify with
the ‘brave heroes’. Which adult would read this classic in such a way
today? And if the form is reproduced faithfully, the strange hexameter
verses will reduce the suspense in any case.

It has been frequently claimed that a translatum should have the ‘same ef-
fect’ as the source text; but, for this to be possible, it may be necessary to
change the function. For these reasons, we shall not regard invariance of
effect as an absolute requirement.