The speech act of promising in an intercultural perspective
1 Some data
2 Locating the cultural misunderstanding
3 To promise or not to promise? Searle’s answer
4 Interpreting the polite promise
5 Politeness and conversational structure
6 The African promise: the outcome of a negotiation
7 Lexical expressions for promising in some West African languages
In the following paper∗ I would like to look at the speech act of promising in an intercultural
perspective. Indeed, my personal intercultural experience in West Africa has seemed to
challenge the assumption that promising is universally understood as a commitment to do
something. I have often been disappointed by people not keeping their promises until I realized
that what I interpreted as a promise may not have been meant to be one in the first place. Then I
started looking for other possible interpretations of what people were conveying when I thought
they were making a promise to me.1 Some data
My admittedly limited source data come from situated conversational exchanges in which I was
either involved myself, or of which I have found reports.
Let me start out with an exchange that I myself participated in. A few weeks before an important
ceremony at SIL Abidjan, I handed an invitation to one of my Ivorian friends working in the
government. When I did so, he said “I shall be there.” However, in the course of the
conversation, he mentioned that on the day of the ceremony he would be six hundred kms away
from the capital on an important government mission. When I pointed this out to him by saying
something like “So then you won’t be able to be there for the ceremony,” his answer was “I shall
be there. I’ll do everything to be there.” As you can guess, he never came. And he must have
known that he couldn’t come when I gave him the invitation. But why did he not express his
regret for not being able to come and thereby excuse himself, instead of making what to me was
a false promise? Those were my questions at the time.
An SIL colleague from Ghana1 reports an even more striking example. Her article describes the
appropriate way, in Koma culture, of expressing thanks for a gift or a service rendered. Not only
will the receiver express his thanks immediately, but he will go to greet the benefactor the next
morning saying “Thank you for yesterday.” Cahill recounts how one day she drove three and a
half hours to take a man and his wife to another village. When she left them there, the man said
he would come the next morning to thank her! Of course he did not come. And he must have
known full well that he wouldn’t. But why did he say so, then, especially since there were
alternate ways of expressing his thanks, which do not necessitate a personal visit, such as “When
you see the sun rising in the morning, then you should know that it is bringing my thanks to
you,” or “When you hear the cock crow in the morning, then you know he is bringing my thanks
to you?”2 Locating the cultural misunderstanding
Making promises that you know you cannot keep seems strange to a Western mind. In Western
culture saying things like “I’ll come to the ceremony” is meant and understood as a promise.
And promising something is committing oneself to doing it.
∗This paper was to be given in Zurich at the International Symposium on “Text in context: African languages
between orality and scripturality”. October 18–20, 2001.
1Ginia Cahill (1992)
Moreover, before promising something, the Westerner wants to be sure s/he will be able to do
what s/he promises. S/he may well desire to do something that s/he knows the hearer would
prefer her/him to do, but unless s/he has reasonable evidence for the fact that s/he is also able to
do it, s/he will not make a promise to do it. S/he will rather say something like “Sorry, I’d like to
do X, but I’m afraid I can’t,” or “I’ll try, but I can’t promise.”
In African culture, however, stating openly that one is not able to do something which one
assumes the addressee expects or prefers one to do is paramount to saying that one doesn’t care
about the relationship with the addressee. It is a terrible loss of face for both the speaker and the
addressee. The way out of this interactional dilemma seems to be to SAY that one will do what
one thinks is preferred or expected by the addressee and, thus, convey that one does care. A
passage from the Senegalese novel Segu2
written by Maryse Condé confirms this in quite a
Amadou Cheikou, King of Macina, dismisses his counsellor Alhadji Guidado who has asked to
withdraw in order to join his son’s wedding.“I’ll come later on,” he
[the King] said, “and join in the young couple’s prayers.” This was
mere politeness. Everyone knew he never went out now.
Here the author, possibly with a Western reader in mind, gives politeness as the key assumption
to the correct interpretation of the King’s utterance, which otherwise contradicts the information
that he never goes out anymore
Thus, at least in some West African cultures, the fact that the speaker thinks a given behavior is
preferred or expected by the addressee seems to be a sufficient condition for what I shall call a
“polite promise” to behave in that way. And s/he seems in no way to feel bound by this
“promise”, nor will anyone expect that s/he fulfill it.
In Western culture, on the other hand, the speaker is bound by a promise. Therefore ability to do
what one promises is considered a necessary requirement for making a promise. Giving a
promise without having the ability to fulfill it constitutes a serious risk to the speaker’s face. So
in order to save face, s/he presents his/her excuses about inability to carry out the action s/he
cannot promise to do.
So for both Westerners and Africans, loss of face is at stake when making a promise, but the
reasons why they lose face are different, and as a matter of consequence, the language behaviour
by which they save face is different, too.What my data and my own intuitive Western reaction
suggest is that Africans will make a polite promise in order to save face, notwithstanding their
ability to actually fulfill it, while Westerners, instead of making a promise that they know they
can’t keep, will rather present excuses, equally in order to save face.3 To promise or not to promise? Searle’s answer
If it is indeed true that in African culture you can make nonbinding polite promises, we need to
ask the question whether these statements are really to be considered as speech acts of
promising. We are, therefore, going to look at the classical definition of the speech act of
promising given in Searle’s Speech Acts
According to Searle, each speech act has four so-called felicity conditions, which make it a
successful act: a condition stating the nature of the propositional content of the sentence
expressing the act; one or several preparatory conditions; a sincerity condition; and an essential
For the speech act of promising, he states the following conditions (A=act; S=speaker;
(1) propositional content: A must be a future act of S
(2) preparatory conditions:
- a promise must be something H wants done or at least would prefer to have done rather
than not done
- S will not do the act in the normal course of events
(3) sincerity condition: S intends to do A
(4) essential condition: S intends that the utterance (T) will place him under an obligation to do
Considering our three examples, the following can be said about each of the felicity conditions
stated by Searle for the act of promise.
(1) The condition of propositional content is satisfied, since in each case the speaker says that he
will carry out some future action.
(2) Both preparatory conditions seem also satisfied, since the action is pleasing to the hearer but
would not happen in the normal course of events.
(3) The sincerity condition is a difficult one as it involves the undefined notion of intention. We
can only speculate about the intentions of the men in the quoted examples. But at a common
sense level, why should the men intend to do something they know already at the moment of
speaking they will not be able to do? And is it possible to STATE an intention without
having it, except in lying? It thus seems more correct to say that the sincerity condition is not
fulfilled for the “polite promises” quoted previously.
(4) The essential condition also involves the problematic notion of intention. However, if the
men do not intend to commit themselves to doing what they say, which seems precisely to be
the case, they will certainly not think that their utterance places them under an obligation to
carry out the stated action. It thus seems fairly safe to say that this condition is not satisfied
either for our data.
According to Searle’s definition, then, none of the speakers in our examples makes a promise at
all! But what is it then that they are they doing?4 Interpreting the polite promise
As my third example from the Senegalese novel showed, politeness is the motivation for the
speaker to make the nonbinding promise and the key assumption in the inferential process
leading to its correct interpretation by the hearer. But how can this inferential process be viewed
and what is the rationale behind this kind of politeness?
One possible account of the inferential process is through Grice’s cooperative principle and his
conversational maxims,which claim to formulate common standards of human communication.
The cooperative principle reads as follows:
“Make your contribution such as is required, at the stage at which it occurs, by the accepted
purpose or direction of the talk exchange in which you are engaged.”(1975:45)
The maxims are concerned with informativeness (maxim of quantity), truth (maxim of quality),
relevance (maxim of relation), and brevity and clarity (maxim of manner) of speakers’
utterances. Politeness requirements in different cultures can easily be shown to violate any of
these maxims. Thus the use of the question form to express a polite request or a polite offer is
considered a violation of the maxim of relevance. Since it is assumed that the speaker still
respects the cooperative principle, which is thus superordinate to the conversational maxims,
additional assumptions, which Grice called implicatures, are necessary to show that the principle
is observed, after all. Politeness is such an implicature.
Brown/Levinson (1987), who build on Grice in their classic treatment of politeness in
communication, make the concept of face; i.e., the individual’s self-esteem, the core concept of
politeness. Face is defined in terms of so-called face-wants, which may be positive (roughly the
desire to feel appreciated) or negative (roughly the desire to remain undisturbed). Depending on
whether the speaker’s politeness strategy aims at preserving the positive or the negative face-
wants of the addressee, Brown/Levinson talk about positive or negative politeness respectively.
In their model, then, “it is the mutual awareness of ‘face’ sensitivity, and the kinds of means-end
reasoning that this induces, that together with the CP [cooperative principle] allows the inference
of implicatures of politeness. From the failure to meet the maxims at face value, plus the
knowledge of face-preserving strategies, the inferences are derived.”(1987:5)
Brown/Levinson mention offers and promises as a positive politeness strategy, as one way of
conveying to the addressee that one is cooperating with him/her. Interestingly enough, they also
admit the possiblity of false promises (1987:125): “even if they are false […] they demonstrate
S’s [speaker’s] good intentions in satisfying H’s [addressee’s] positive-face wants.”
That is exactly what we have in our data, and therefore the solution to our question what the men
are actually conveying in making polite promises.
Given the fact that relationships are all important in African culture, it is to be expected that
politeness will play a preponderant role in all verbal interactions. Maintaining good social
relationships is traditionally a matter of social and even physical survival in Africa. Preferred
ways of maintaining good relationships are giving, including doing favors, and visiting. Not
giving and not visiting constitutes a risk to a good relationship. Face-threatening acts like
refusals and criticisms are generally avoided altogether, because they constitute too great a risk
for the relationship.
So in order to show that he wants the good relationship to continue, the man in the example from
Ghana said he would come to greet in the morning, even though he knew he wouldn’t come.
Such behavior must seem insincere to a Western mind but is possible in a culture where the
desire to maintain a relationship has priority over sincerity. It is acceptable in Koma culture to
state the intention of coming to say thank you the next morning as required by the culture while
being unable to do so under the given circumstances.5 Politeness and conversational structure
The bulk of studies on politeness phenomena in language use concentrates on how politeness is
manifested at the level of individual sentences, such as through the use of certain modal
particles, honorific pronouns, subjunctive modes, tenses, or certain syntactic structures.
However, as one looks at whole conversational exchanges or even encounters in face-to-face
interaction, rather than just at the sentence realizing a certain speech act, it appears that
politeness has also to do with the way encounters are opened, carried through, and closed.
In many African cultures, elaborate greeting rituals consisting of a whole chain of exchanges
testify to the fact that in an encounter the appropriate management of the social relationship is as
important as the content which will be spoken about during the encounter.
In the closing of an exchange or an encounter, the social relationship comes again to the fore. In
Goffman’s interactionist terms, the exchange is closed when the interactional balance is
reestablished. In the framework of the Geneva model of conversational analysis, Roulet et al.
(1985) have taken up this idea of Goffman’s in formulating the structural constraint of double
agreement for the successful closure of a conversational exchange. This constraint stipulates that
an exchange cannot be completed unless both parties agree, which means that the last two
moves3 of the exchange must have the same argumentative orientation. It becomes immediately
obvious that an exchange cannot satisfactorily be closed by a negative move like a refusal.
In the examples quoted previously, the conversational moves expressing the false “promises” are
all reactive moves that allow the closing of the verbal exchange as normally expected. In the
first example, my friend reacted positively to my invitation, thereby allowing the exchange
about the invitation to be closed. In the second example, the Ghanaian Koma man reacted
verbally to a nonverbal act; namely, to a service rendered to him and his wife. In the third
example, the King reacts to his counsellor’s request for dismissal.
The “truth” in each of these cases would have resulted in some kind of a negative move, such as
“I have some other obligations,” “I won’t be able to be to be there,” which would not have
allowed the exchange to be closed satisfactorily at the interactional level. In fact, when, in the
conversation with my friend, I reiterated the exchange about the invitation later in the
conversation by pointing out that he had said he wouldn’t be able to follow the invitation, he
only reiterated his promise in order to quickly allow the exchange to close again. He was
obviously unwilling to engage in a negotiation by taking my remark into account, which
incidentally was very face-threatening to him!
Interestingly enough, in the example from Ghana, there were culturally acceptable alternative
utterances which did not imply a personal visit and would also have allowed the exchange to
close satisfactorily in structural terms. Nevertheless the man chose to “promise” the impossible
visit. The explanation for this choice seems to be that the man’s thankfulness towards the white
woman was so great that he did not feel that the alternatives would express it adequately and
rather chose to make what to a Western mind is a “false” promise.
What I have suggested in this section, then, is that nonbinding promises can be used in African
languages to politely close a conversational exchange.6 The African promise: the outcome of a negotiation
Being now able to interpret the polite promise in African culture, we still need to ask whether
there is a way to promise in the Searlian sense; i.e., an utterance where the speaker commits
himself/herself to an action and is also understood by the addressee as doing so. I am only
3An exchange consists of three moves: initiative, response, evaluation.
considering promises in everyday situations, not formal or ritual promises, which are normally
accompanied by some gesture and have heavy sanctions if broken.
When I asked a young Ivorian about how, in an informal situation, one could be sure of the
commitment of the other person, I got the following quite revealing response: “If somebody tells
you that he will do something, this is not yet a promise. But you may ask him back, then he will
tell you if he is serious.” For speech act theory and in terms of conversational structure, this
means that the speech act of promising is not realized in just one conversational move but
through an exchange. Searle’s essential condition; i.e., the commitment is thus the outcome of a
negotiation, rather than being achieved by a single sentence.4 Unfortunately, I have not yet been
able (or attentive enough) to observe such a negotiation in a real life situation, but I found an
example in a Kenyan novel (Ngugi 1969:96).
Girl: When I come back, you will not let me alone?
Boy: When you come back, I shall be with you.
Girl: It is a promise?
Boy: Of course.7 Lexical expressions for promising in some West African languages
Before concluding, I would like to look at some lexical expressions that may accompany a
binding promise in some West African languages.5
In Nyaboa, a Kru language of Côte d’Ivoire, the expression used for a binding promise is “I put
it into your mouth.” This performative utterance follows the utterance expressing the promise. A
promise then sounds like this “I’ll come to the funeral; I put it into your mouth.” In reporting this
promise, the hearer will say “S/he put it into my mouth saying s/he will come to the funeral.”
In Krumen, also a Kru language of Côte d’Ivoire, there seems to be no performative verb for the
act of promising. From Godié or Wobé, two other Kru languages, I do not know of a
performative verb for promising either.
However, in reporting a binding promise, the Krumen speaker has the possibility of adding the
verb “throw” before the speech introducer lee
‘say’, e.g., “He threw lee
he will come to the
funeral.” When a Wobé makes a binding promise, he seems less concerned about his own
obligation than about reassuring the hearer. He will say, “Trust me, I will do X.”
4Brown/Levinson (1987:10) address precisely this criticism to speech act theory: “speech act theory forces a
sentence-based, speaker-oriented mode of analysis, requiring attribution of speech act categories where our own
thesis requires that utterances be often equivocal in force.” Along similar lines, Saville-Troike (1982:36) reports
how in Indian culture, the realization of an individual speech act like that of accepting an offer spread over a whole
5My main source of information here has been a questionnaire with eight sentences in French to be translated in the
various languages. The sentences are the following: (1) I will be there for the funeral. (2) I promise you to be there
for the funeral. (3) S/he told me/us that s/he would come to the funeral. (4) S/he promised me/us to come to the
funeral. (5) S/he promised me she would find work for me in Abidjan. (6) S/he makes promises but doesn't do
anything. (7) God has promised us a new heaven and a new earth. (8) God does what he promises.
My thanks for giving information go to Toto Nemlin Laurent (Krumen), Robert Carlson (Supyire), Georges Kodjo
(Nzema), Kouassi Tanoh (Baoulé), Kéhi François (Wobé), Julie Bentinck and Raoul Fran (Nyaboa), Tan Deli
Antoine (Dan), and Mathieu Ouattara (Djimini).
The verbal phrase “pick a debt” in Dan, a Southern Mandé language of Côte d’Ivoire,
emphasizes the obligation aspect of the promise. The phrase does not have a performative use,
however, but only a reportive use. Without using the phrase “pick a debt,” the speaker may
emphasize that the promise is binding for him by adding the performative expression “I say in
truth” at the beginning of his utterance or else only the adverbial “in truth” or other adverbials
meaning “truly, certainly” at the end of his utterance. A modal particle instead of the adverbials
at the end of the utterance also strengthens the promise.
The idea of the promise as a debt is also explicit in Djimini, a Senufo language of Côte d’Ivoire.
The expression “to cut a mouth-debt in regard to someone” is used performatively in making a
promise as well as for reporting a promise. An alternative expression is “to give into someone’s
In Supyire, a Senufo language of Mali, there are also several lexical expressions for the binding
promise. “I promise to you that X” is translated either by the verbal phrase “I exchange my
mouth to you (I say) X,” X always being a complement clause with simple future tense, not
subjunctive or infinitive.
The expression “I exchange my mouth to you” is also at the basis of the noun compound
“mouth-exchange,” meaning “promise.” Using this noun, another expression is “I give my
mouth-exchange to you.” All of these expressions refer to a conversational exchange.
Another noun with the meaning of “promise” in Supyire is “mouth-rope,” referring to the
binding character of the given promise. The sentence “S/he doesn’t keep her promises” is
translated as “S/he gives a mouth-rope, but she doesn’t grab it.” The sentence “God keeps his
promises” is translated as “God goes in the footsteps of his mouth-rope,” or “God does his
Interestingly enough, the noun compound “mouth-rope” can also have the meaning of “order” in
a sentence like “S/he grabbed my mouth-rope” for “S/he obeyed my order.” Maybe the better
translation for the expression “mouth-rope” is “obligation.” Indeed, an order places the hearer
under an obligation to act, whereas a promise does so for the speaker. The owner of the “mouth-
rope” is thus the author of an obligation under which he places either the hearer (in the case of
an order) or himself (in the case of the promise).
Baoulé and Nzema, two Kwa languages, are the only languages among the ones about which I
have information where a binding promise is expressed by a simple future I-statement like in
Western languages. Both informants from these languages affirmed that such a statement implies
an obligation for the speaker. In Baoulé, the verb “wan,” whose basic meaning is ‘to speak,’ can
only be used by people who are important and have the means to keep their promise. It is the
word invariably used for God’s promises.
But what about the polite promise in these languages, then? Does it exist? Yes it does! The
Nzema language has utterances with linguistic clues to say that the promise is not binding. If
someone says “I shall try to X.” or “I shall try to find the means to do X,” the hearer knows by
conventional implicature that the speaker has no intention to X, but that he is being polite. The
Nzema example shows that even in African languages where at first sight the binding promise is
expressed by a simple I-statement like in European languages, provision is made for the
In conclusion, then, Searle’s definition of the promise can be basically maintained for the act of
promising in Africa. However, the African promise differs from the Western one in the way it
comes about; i.e., by negotiation of the felicity conditions. On the other hand, for an African
speaker, just stating an intention to perform an act in the hearer’s favour does not yet imply
commitment or even presuppose ability to carry out the act. In fact, statements of this kind are
generally to be interpreted as polite promises; i.e., ways to satisfy cultural expectations and save
face. Intercultural misunderstandings in relation to this type of statement of intention arise if
satisfaction of all the felicity conditions for the act of a “classical” promise are assumed to be
fulfilled each time such a statement is made.
The examination of lexical expressions from some languages has shown that the type of promise
by which the speaker puts himself under an obligation exist in all of them.References
Brown, P., and S. C. Levinson. 1987. Politeness. Some universals in language usage.
Cambridge University Press.
Cahill, G. 1995. “Thanking in the Morning.” In H. Hill and J. Arensen (eds.), The Best of Ethno-Info 22
Summer Institute of Linguistics, 38–39.
Condé, M. 1987. Segu.
New York: Ballantine books.
Egner, I. 1995. “Politeness in Language: The Case of Interrogatives.” In H. Hill, and J. Arensen (eds.), The Best ofEthno-Info
. Nairobi: Summer Institute of Linguistics, 55–57.
Egner, I. 1996. “Other-repetition in Question Form: Evidence from a West-African Language.” In C. Bazzanella
(ed.), Repetition in Dialogue
. Tübingen: Niemeyer, 66–77.
Goffman, E. 1971. Interaction ritual
Grice, H. P. 1975. “Logic and Conversation.” In P. Cole, and J. L. Morgan (eds.), Syntax and Semantics 3: SpeechActs
. New York: Academic Press, 41–58.
Ngugi, Wa Thiongo. 1969. Weep Not, Child.
Roulet et al. 1985. L’articulation du discours en français contemporain.
Bern: Peter Lang.
Saville-Troike, M. 1982. The ethnography of communication.
Oxford: Basil Blackwell.
Searle, John R. 1969. Speech acts.
London:Cambridge University Press.
- 0 Introduction
- 1 Some data
- 2 Locating the cultural misunderstanding
- 3 To promise or not to promise? Searles answer
- 4 Interpreting the polite promise
- 5 Politeness and conversational structure
- 6 The African promise: the outcome of a negotiation
- 7 Lexical expressions for promising in some West African languages
- 8 Conclusion