Available online at www.sciencedirect.

com

Journal of Pragmatics 44 (2012) 980--996

www.elsevier.com/locate/pragma

‘‘Was that a compliment?’’ Implicit compliments in

English and Spanish
Carmen Maíz-Arévalo *
Universidad Complutense de Madrid, Facultad de Filología, Departamento de Filología Inglesa I, Ciudad Universitaria s/n, 28040 Madrid, Spain
Received 19 September 2011; received in revised form 31 March 2012; accepted 13 April 2012

Abstract
Compliments are certainly one of the most widely studied speech acts, as shown by the extensive literature devoted to the topic.
However, as pointed out by different authors, research has almost exclusively focused on formulaic compliments while neglecting implicit
ones, most probably because the first tend to outnumber the latter. The present paper is an attempt to redress the balance in favor of
implicit compliments. My focus, however, is not on the interpretive procedures employed by interlocutors when receiving an implicit
compliment but on the pragmatic motives underlying the speaker's choice of this much rarer speech act. Thus, this paper aims to answer
the following research question: What are the speaker's pragmatic motivations to opt for an implicit instead of an explicit compliment?
Results show that implicit compliments are preferred when evaluating someone's qualities, achievements or personal appearance in
order to avoid face-threat, especially when the relationship between the interlocutors is still distant. Finally, it can also be argued that not
all implicit compliments share the same degree of implicitness. In fact, some of them show recurrent linguistic patterns which lead to us
argue in favor of a cline of explicitness both in English and Spanish.
© 2012 Elsevier B.V. All rights reserved.

Keywords: Implicit compliments; Pragmatic motivations; Spanish; English

1. Introduction
As Pomerantz (1978) has pointed out, the use of implicit compliments is a helpful way to avoid the dilemma posed by
compliments. This is particularly true of responses to compliments, where there is frequently a clash between the need to
be modest and the wish to avoid disagreement with the interlocutor, which can render responses to compliments a
particularly tricky issue (Maíz-Arévalo, in press). Paying a compliment can also be quite face-threatening for the speakers
themselves since the interlocutor receiving it might not take it as a positive remark to be appreciated but as an unwelcome
or embarrassing comment, as happens in example (1) below, where (M) tries to be polite by saying something positive
about R's appearance, who is clearly embarrassed as his attempt to change the topic shows1:

Example 1:

M: Te veo diferente
‘You look different’
R: hmm, bueno, supongo que más viejo.
‘Hmm, well, I guess I look older.’

* Tel.: +34 5001506.

E-mail address: [email protected].
1
It is difficult to determine the real causes that provoke R's embarrassment, whether it is the gender difference (M is female and R is male),
power differences (R used to be M's boss), intercultural pragmatic differences (M is Spanish and R is British) or a combination of all three. See
Maíz-Arévalo (in press) for possible reasons regarding responses to compliments.

0378-2166/$ -- see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.pragma.2012.04.004
 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996 981
?
M: no, no, qué va, más joven. Ya llevabas barba antes?
‘No, no, no way, younger. Did you use to have a beard?’
R: ejem, esto, sí, bueno, si esto se puede llamar barba [pause] eh, bueno, parece que la charla está a punto de
empezar. . . vamos, vamos.
‘Well, er, yes, well, if this can be called a beard [pause]’, huh, well, it seems the talk is about to start. . . let's go.’
M: sí, vamos, vamos.
‘Yeah, let's go, let's go.’

Following Brown and Levinson (1987), it can be stated that formulaic compliments involve going ‘‘bald on record’’2
whereas an implicit compliment -- as its very name suggests -- implicates going ‘‘off record’’ or hinting. As these authors
suggest, both strategies have advantages and disadvantages. Whilst going bald on record seems to avoid ambiguity and
entail honesty and directness on the speaker's part, for the very same reasons it can also threaten the hearer's face in certain
circumstances (e.g. if the relationship between the interlocutors is distant). On the other hand, choosing the off- record
strategy poses the problem of risking not being correctly understood, which could be a face-threat for the speaker. In this
case, however, the hearer's face would be more ‘preserved’. This might lead us to think that implicit compliments should have
a higher frequency in casual conversation than they actually do (especially in contexts where the relationship between the
interlocutors is not particularly close). However, this is not the case and formulaic compliments are by far the most frequently
used, as proved by the extensive research both on English and other languages like Chinese, Japanese, Polish, Spanish,
German, French and so on3 (Bruti, 2006; Chen, 1993; Golato, 2002, 2005; Herbert, 1989, 1990, 1991; Herbert and Straight,
1989; Holmes, 1986, 1995; Jaworski, 1995; Kerbrat-Orecchioni, 1987; Knapp et al., 1984; Lewandowska-Tomaszczyk,
1989; Liu, 1995; Lorenzo-Dus, 2001; Plasencia and Yépez Lasso, 1999; Pomerantz, 1978; Saito and Beeken, 1997;
Wolfson and Manes, 1980; Wolfson, 1981, 1983; Ylänne-McEwen, 1993; Yuan, 2002 among many others).
One reason for this preference of formulaic compliments over implicit ones might be that face-threat is considered less
determining in the choice of how to pay a compliment than the risk of not being correctly understood by the interlocutor if
our compliment is too indirect. The aim of the current article is thus to ascertain the interlocutors’ pragmatic motivations for
preferring such a ‘‘marked option’’ (i.e. paying an implicit compliment instead of a formulaic one) in Spanish.4
The paper is divided into six parts. Section 2 below focuses on describing the methodology and the data used in the
study. Section 3 briefly reviews the literature on implicit compliments, also defining what makes a compliment implicit as
opposed to explicit, more formulaic ones. The fourth and fifth sections discuss results and explore the possible
motivations underlying the speaker's choice of one type of compliment instead of another. Finally, section 6 offers the
conclusions and points to future research.

2. Data and methodology

There have been different methods of gathering data by authors researching compliments, the most common ones being
Discourse Completion Tests (DCTs), questionnaires, role plays and interviews. As Golato (2003, 2005) points out, these
methods have the advantage of allowing researchers to collect a great deal of data in a relatively short period of time as well
as to control certain variables such as age or educational background (among others) since participants can also be selected
if need be. However, the data they render are usually prototypical samples of the language under study, being thus unlikely to
provide less prototypical examples as implicit compliments seem to be. This is why both Boyle (2000) and Golato (2005)
argue for the use of audio- or video-recording of naturally occurring data, ‘‘capable of producing a much less formulaic
collection of compliments’’ (Boyle, 2000:29). Their claim is also supported by other studies like Bruti (2006), who does not
employ naturally occurring data but scripts in English whose Italian translation reproduces more or less loyally ‘‘real
conversations’’. Boyle's study is a pioneer attempt to analyze implicit compliments in English by focusing on the mechanisms
by which interlocutors correctly interpret implicit compliments as such. However, he pays less attention to the pragmatic
motivations underlying the interlocutors’ choice of implicit compliments, which is my main objective in the present paper.
This paper follows Boyle and Golato in their ethnomethodological perspective since I agree that compliments are
discursive and should be observed within naturally produced conversations. Thus, the data employed in the current

2
In Brown and Levinson's theory, going baldly on record consists of stating explicitly the illocutionary force of the speech act. This explicitness
can then be redressed by means of politeness strategies (either positive or negative). It is beyond the scope of this paper to revise the
weaknesses of this theory (e.g. the overlapping of maxims, the simultaneity of positive and negative politeness strategies in the same act and so
on). Suffice it to say, however, that I consider their distinction between going bald on record and off record very suitable and appropriate for the
current analysis.
3
In his paper on implicit compliments, however, Boyle (2000:28) argues that it is not the fact that explicit compliments are much more frequent
than implicit ones but that the methodology commonly employed and the restricted view of the phatic function of compliments makes implicit
compliments much harder to identify than formulaic ones.
4
The variety under study is Peninsular Spanish.
982 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996

analysis all proceed from a corpus of seventy non-elicited audio-recorded conversations in Peninsular Spanish.5 Apart
from personal reasons (see note 5), the choice of Spanish is also due to the fact that, up to date, studies in this language
(and its varieties) have focused on formulaic, explicit compliments while neglecting implicit ones (Valdés and Pino, 1981;
Moore, 1996; Lorenzo-Dus, 2001; Maíz-Arévalo, 2010; Plasencia and Yépez Lasso, 1999). To my knowledge, there is no
study of implicit compliments in Peninsular Spanish.
The recording period spanned throughout the last three years and, in order not to bias the data, the recordings were
carried out randomly, without knowing what the topic(s) might be or what participants would be involved. With regards to
the participants, it should be stated that the researcher was always involved in these conversational exchanges some way
or another (i.e. as a speaker or as one of the hearers), which helped clarify certain ambiguities and ensure as much as
possibly that an implicit compliment was taking place.6 In some examples, meta-pragmatic comments helped clear up any
ambiguities.7 Furthermore, as a follow-up, six of the participants involved (three male and three female interlocutors) were
also interviewed8 so as to gain a deeper and more objective insight into the pragmatic motivations to pay an implicit
instead of an explicit compliment. These interviews were crucial to ascertain the participants’ impression of their own
words and intentions behind them. As Golato (2005:24) points out, ‘‘the participants in the interaction have put
significance on that particular event -- that is, the participants have demonstrated that the event is relevant to them’’.
As for the recordings, all the data were surreptitiously gathered in order to avoid ‘‘unnatural’’ conversation, but the
participants were afterwards asked for permission to reproduce their conversations. With regard to the transcription of the
exchanges, it has been kept simple for the sake of clarity. Following Halliday and Hassan's notion of ‘‘context of situation’’
(1976), each example in the study is preceded by as much contextual information as possible about the field (what is going
on) and the tenor (who are the participants and their relationship). The mode (kind of language used, oral or written, etc.)
has not been taken into account since all the exchanges are oral. As has probably been inferred by the reader, the
approach adopted throughout the paper is unabashedly functional. In other words, it is firmly believed that form cannot be
separated from the function(s) it fulfils.9
All in all, over four hundred compliment exchanges were collected (including both formulaic and implicit compliments).
Quite remarkably, more than 90% of these were formulaic compliments,10 which were analyzed in depth in previous
studies from a quantitative and qualitative approach (Maíz-Arévalo, 2010, in press).
In her opinion, formulaic compliments open up a network of linguistic realizations suitable for the expression of different
functions (e.g. expression of emotions through exclamative sentences). On the other hand, implicit compliments in their
system are not only less frequent but also unpredictable as far as their linguistic realization is concerned. However, implicit
compliments do reveal a set of concrete realizations as will be shown in the following sections. Indeed, it might be argued
that they are not that implicit after all. Following Searle's notion of ‘‘conventionally indirect speech acts’’ (1969), most
implicit compliments could be redefined as conventionally implicit since they also follow a system of choices, as will be
explained later.

3. Theoretical background

3.1. Explicit versus implicit compliments

Compliments have been defined as speech acts that ‘‘[. . .] explicitly or implicitly attribute credit to someone other than
the speaker, usually the person addressed, for some ‘good’ (possession, characteristic, skill, etc.) which is positively
valued by the speaker and the hearer’’ (Holmes, 1986:485). According to Boyle (2000:18), compliments can be
considered explicit when they ‘‘are recognized as compliments outside of context, being realized by a small set of
conventional formulae’’. Leaving aside other context-bound issues such as irony or sarcasm, any ordinary speaker would

5
The fact of living in Spain makes it quite difficult for the author to record spontaneous conversations among English native speakers. Hence,
we follow Boyle's examples in English for contrast. Some of Bruti's examples (2006) are also used, even if they come from scripts rather than from
natural conversation, since they serve to illustrate the same phenomenon.
6
In this respect, we follow Golato's study of German compliments (2005, 2006), who also employs a ‘personal’ corpus of examples where she is
usually a participant. This ensures the background knowledge shared by the interlocutors is also well known to the researcher, which might clarify
certain ambiguities.
7
We are referring here to cases where the hearer, unsure about the illocutionary force of the speech act follows up with a clarifying request such
as ‘‘what did you mean by that?’’ or where the speaker clarifies afterwards with comments such as ‘‘I meant it as a compliment’’.
8
All the interviewees are non-linguists. Far from being a disadvantage, my belief is that they are truly speaking their mind and unbiased and
following their intuitions as to why they behaved the way they did. Participants were reminded of the whole conversational exchange (field and
tenor). Then, they were requested to read the transcription of their own utterances. After that, they were asked to explain why they had spoken that
way and not any other and what their communicative intention(s) were.
9
More concretely, we follow Halliday's Systemic Functional Grammar, especially his notion of ‘‘systems’’ or networks of choice for speakers.
10
The number of implicit compliments in this corpus amounts to 40 compliment exchanges (less than 10% of the total corpus).
 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996 983

in fact recognize the following expressions as compliments because of their linguistic realization: ‘‘I love your shirt’’ or
‘‘what a lovely dress you’re wearing today!’’
Explicit compliments are thus linguistically realized by declarative sentences which in turn can be affirmative or
exclamative.11 There is also a limited lexical choice of verbs (gustar, encantar, love, like) and adjectives (bonito, lovely), as
the extensive bibliography on the topic has shown (Boyle, 2000; Herbert, 1989, 1990; Herbert and Straight, 1989; Holmes,
1998; Knapp et al., 1984; Pomerantz, 1978; Wolfson and Manes, 1980; Wolfson, 1981, 1983 just to quote some of the
studies that have focused exclusively on English varieties).
Thus, the analysis of Spanish formulaic compliments renders three major lexico-grammatical patterns, as follows
(Maíz-Arévalo, 2010):

Exclamative clause with an adjective group modifying complimented item. For example: ‘‘ Qué preciosidad
!
i.
de pulsera!’’ (‘What a beautiful bracelet!’).
ii. Declarative clause with the complimentee in the subject position, followed by a copulative verb and a positive
adjective. For instance: ‘‘eres un bellezón’’ (‘you are a real beauty’)
iii. Declarative clause with the complimenter in the subject position, followed by a mental process of ‘liking’ and
the complimented item in the direct object position such as: ‘‘me encanta tu camisa’’ (‘I love your shirt’).

This has led Maíz-Arévalo and García-Gómez (forthcoming) to argue in favor of the following system of compliments
[(Fig._1)TD$IG]
(Fig. 1).

Fig. 1. System of compliments (Maíz-Arévalo and García-Gómez, forthcoming).

Implicit compliments, on the other hand, are those ‘‘in which the value judgement is presupposed and/or implicated by
Gricean maxims’’ (Boyle, 2000:28). In other words, they do not necessarily have a ‘‘fixed’’ linguistic form (although see
section 4) like explicit compliments do and hearers need to infer the corresponding implicature for their interpretation.
Boyle (2000:28) gives the following examples to illustrate implicit compliments in English: ‘‘I wish I could manage my work
like you do’’ and ‘‘Your husband is a very lucky man’’. While I agree that both examples are implicit compliments, they are
relatively formulaic too, especially the second one, which also occurs -- in exactly the same form -- in our Spanish data:

Example 2:

[Field: Two work colleagues are saying farewell after a celebratory lunch together. C has told K that she is engaged.//
Tenor: K (male, around 45) and C (female, 38) are not close friends but their working relationship is developing into
friendship. However, they still have not met outside work.]

K: Oye, lo dicho, que me alegro mucho de lo de tu boda. Tu prometido es un hombre muy afortunado.
‘Listen, let me just tell you again, I’m really happy about your wedding. Your fiancé is a very lucky man.’
C: Hombre, muchas gracias.
‘Man, thanks a lot.’

11
Although it is true that these linguistic patterns account for practically all explicit compliments (Maíz-Arévalo, 2010), there are other linguistic
realizations such as elliptical cases where only a nominal group is employed (e.g. Menuda boca! ‘That's some mouth!’). This is particularly the
!
case in online compliments (Maíz-Arévalo and García-Gómez, forthcoming).
984 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996

On other occasions, however, implicit compliments are so ‘‘obscure’’ that they are much more difficult to be interpreted
as such, which is why addressees go to the extent of ‘‘asking for help in the interpretation’’, as in the following case
retrieved from the Internet12:

Example 3:
(2) [Question] Was this a compliment? Someone wrote to me in spanish.? (sic)
I was on Facebook then someone wrote to me ‘‘Well hello there actora de telenovelas;) <3’’ he said thinks I speak Spanish
because I am latina but i don’t speak spanish. Lol anyways was that a compliment? Actora se telenovelas? (sic) I know it
means ‘‘soap actress’’ but that's weird but then again I know nothing about Spanish slang.
[Answer 1] I would think it's a compliment, maybe they were saying like ‘‘Hey there movie star’’ which would imply that you
are attractive. Just a guess [chosen as best answer]
[Answer 2] It's rather ambiguous, but then again soap actresses are over the top. A lot would depend on what else he said,
of course. Since he didn’t explicitly call you maja or hermosa or guapa or linda, I would suggest you simply take it as a
compliment and think no more of it.
[Answer 3] it was meant to be a compliment.

4. Results

According to Boyle (2000:35), there are two main ways in which people pay implicit compliments in English. The first
‘‘involves comparing a person to someone that the speaker thinks the other person might admire’’; the second ‘‘involves
referring to something that a person has done and that he or she is proud of’’. As already pointed out, one way in which we
can compliment our interlocutor is by comparing them to someone worth admiring. Boyle (2000:40--41) illustrates this
phenomenon with the following two examples, which we partially reproduce here13:

Example 4:
JL: There's there's one very Andrews Sisterish track
KDL: Smoke Dreams smoke dreams yeah (.) (laughs) thank you oh oh

Example 5:

JL: Lovely arrangements on the album though ahm (pause) there's something that I hadn’t really spotted before in you that
comes out on this album there's something Karen Carpenterish about your voice on this album, is that deliberate?
Kdl: No, it's not deliberate but I have heard that comparison before ahm I I guess it would be you know the tone of voice and
maybe the vibrato but (.) you know I take it as a big compliment because I am a huge Karen Carpenter fan
JL: well that's nice phew (laughs) I thought you might be worried about it
Kdl: no no
JL: I think she's a fantastic voice Karen Carpenter
Kdl: yeah (it) is yeah

In both cases, JL (a radio interviewer) chooses to pay an implicit compliment to his guest Kdl (a Canadian singer). As
shown by her response, Kdl immediately recognizes (and accepts) it as a compliment in example (4). This is paralleled by
the following case in the Spanish corpus:

Example 6:
[Field: Two work colleagues are talking after C has given an important talk that she has been preparing for a long time.
E was part of the audience. // Tenor: C (female, 38) and E (female, 34) are not close friends but their working relationship is
developing into friendship. However, they have met outside work only occasionally. J is a senior professor of their
Department that they both admire.]
!
C: Uf, qué nerviosa estaba! Me temblaban las manos y todo.
‘Oh, how nervous I was! My hands didn’t stop trembling’
? !
E: Nerviosa? Pero si me has recordado a J!
‘Nervous? You reminded me of J!’

12
http://answers.yahoo.com/question/index?qid=20110305103546AANItaZ.
13
To keep transcription of examples homogeneous, all the diacritic symbols present in the original have been avoided.
 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996 985
? !
C: Sí? Venga ya! (smiles), jo gracias.
‘Did I? Come on! (smiles), oh thanks’
E: Hala, ahora a celebrarlo con unas cañitas (change of topic)
‘Now, let's go celebrate with a couple of beers’

As illustrated by the contextual information above, example (6) is part of a conversational exchange between two
female work colleagues. C had to give an important talk which was quite successful, as E comments later. The exchange
took place just after the talk. E compliments C implicitly by comparing her with someone they both admire, J. For the
compliment to be correctly inferred by the hearer, it is necessary that C also knows the object of comparison and shares
the speaker's admiration. Taking a term from traditional pragmatics, these would be the ‘felicity conditions’ that guarantee
the implicature to be successfully interpreted and for the compliment to be accepted as such. C's response in turn
(3) shows that she has correctly interpreted E's utterance as a compliment.14
In these cases the comparison is clearly felicitous and both interlocutors recognize the object of comparison as admirable.
Quite remarkably, both the English and the Spanish examples seem to share another main feature: the complimentee's
slightly embarrassed giggle preceding further appreciation, which shows the remark was modestly accepted.15
Example (5), on the other hand, shows a more elaborate negotiation where Kdl finally ‘‘takes it as a big compliment’’. JL's
following relief (‘‘well that's nice phew’’) shows he was well aware of being taking a risk since Kdl might not even know or like
the object of comparison, that is, she might even be offended as in the following example from our Spanish corpus (7):

Example 7:
[Field: Two female friends are talking about B's new haircut (with new fringe included). // Tenor: A (female, about 30) and B
(female, about 25)]
!
A: Vaya, te has cortado el pelo. Pareces Cleopatra!
‘Wow, you’ve had a haircut. You look like Cleopatra!’
B: [pause] Bueno, pues ya crecerá, tú tampoco lo llevas muy allá.
[pause] ‘well, it’ll grow back, won’t it? You don’t look so cool yourself ’
A: [pause] Esto, bueno, yo lo decía como algo bueno, un cumplido, vaya. Joder.
‘[pause] Oh, well, I meant it as something nice, a compliment, oh f ***.’
B: Ah, esto, vaya corte. Vale, pues la próxima vez habla más clarito y ya está.
‘Huh, well, this is embarrassing. Well, next time speak more clearly and that's it!’

In this exchange between two female friends, the interlocutors do not share the same appreciation of Cleopatra's haircut
as a model of beautiful hair. Hence the implicit compliment fails to make its point, as shown by the meta-pragmatic comment
made by A (in bold) and the response it gets from B (underlined). The following three examples from two conversational
exchanges16 in the corpus also show comparisons which are partially infelicitous for different reasons. Thus, in example (8),
the object of comparison is unknown to the hearer, who looks puzzled in response for a clarification. This comes as an explicit
compliment in S’ following turn, which is undoubtedly accepted as such by B as shown by her grateful smile:

Example 8:
[Field: Some friends are talking about B's new haircut (with new fringe included). // Tenor: B (female, about 25) and S
(female, about 30)]

S: Así te pareces mucho a una amiga mía

‘You look like a friend of mine’
B: (looks puzzled)
S: Bien guapas estáis las dos
‘You both look very pretty’
B: (smiles)

14
As shown in a previous study (Maíz-Arévalo, in press), C's formulaic ‘‘anda ya’’ (no way!) is a typically response to compliments in Peninsular
Spanish. After this initial rejection, the compliment is accepted and appreciated.
15
The complimentee's giggle in example (6), however, seems to point out that the compliment is somehow embarrassing. It could be argued that
embarrassment might have been higher had the compliment been paid directly. This might be highly difficult to know or prove were it not for the
fact that it was addressed to the present author.
16
Examples (8) and (9) belong to the same conversational exchange. However, they are presented separately for the sake of clarity.
986 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996

Example (9) illustrates a case where the possible infelicitous comparison is explicitly clarified by the speaker in the
same turn, after a short pause. Given that Lady Gaga's attire is admired by some and criticized by others, Sa is unsure
whether S shares his admiration for the singer and quickly makes the comparison explicit:
Example 9:
[Field: Some friends are talking about B's new haircut (with new fringe included). // Tenor: B (female, about 25) and Sa
(male, late 20s)]

Sa: Te lo has cortado como Lady Gaga (pauses and smiles). Te queda muy bien.
‘You have had it cut like Lady Gaga's. It really suits you’
B: (smiles)

Finally, examples (10) and (11) illustrate another comparison which gives rise to the hearer's meta-pragmatic comment
asking for clarification so as to correctly interpret the speaker's compliment:
Example 10:

[Field: C, who is wearing a new pair of glasses, is greeted by a work colleague. // Tenor: E (male, about 45), C (female, 38)
are work colleagues. Their relationship is relatively distant since they meet only from time to time and always at work.]

E: Anda, con esas gafas te pareces a Nana Moskouri.

‘Oh, with those glasses you look like Nana Moskouri’
?
C: Y eso es bueno o malo?
‘And that is good or bad?’
E: no mujer, es bueno. A mí me parece una mujer guapísima y súper elegante.
‘No, dear, it's good. I think she is a very pretty and super elegant woman’
!
C: Vaya, entonces, muchas gracias!
‘Wow, then, thanks a lot!’

Example 11:
[Field: Four friends are having a relaxed drink together. The conversation at this point is purely interactional and focuses on
TV programs they have recently seen // Tenor: Ca (male, 38), C (female, 38), J (male, 38) and Jo (male, in his early 40s). C,
Ca (who is engaged to C), and J have a close relationship whereas Jo is relatively new to the group. He knows C better and
partially Ca but it is the first time he meets J. Ca and C are trying to set them up although they are not aware of this. Despite its
length, the example is reproduced in its totality to show how interlocutors negotiate the infelicitous compliment.]
?
C: No, no lo vimos, estábamos cenando fuera. Qué tal estuvo?
‘No, we didn’t see it, we were having dinner out. What was it like?’
J: Muy bien, era como un documental.
‘Very good, it was like a documentary’
Ca/Jo: (both of them nod)
C: Anda, para documentales el que vi ayer después de comer, por cierto, me acordé mucho de ti porque era clavadito a ti.
‘Oh, about documentaries, I saw one yesterday after lunch which reminded me lots of you, he so looked just like you.’
?
J: de qué era?
‘What was it about?’
C: de Tutankamón . . . (C cannot finish her turn because she gets interrupted by Ca)
‘About Tutankhamon. . .’
Ca: (laughs) Joder, ya te vale a ti también, menudo comienzo (laughs again)
‘F ***, well done, what a nice beginning’
Jo: (smiles shyly)
J: (slightly offended) a ver, guapa, ahora nos lo explicas
‘Let's see, pretty face, now you’ll have to explain that to us’
C: (embarrassed) Jo, cómo sois, es que no me habéis dejado ni terminar. . . me recordó a ti porque, bueno, hacían una, a
ver (pauses)
‘Oh, my, you are, you haven’t even let me finish. . . it reminded me of you because, well, they did, see. . .’
?
J: Sí?
‘Yes?’
 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996 987

C: Pues eso, que hacían una reconstrucción de como murió (pauses and J interrupts her)
‘Well, that, they recreated how he had died. . .’
?
J: Pues sí que lo estás arreglando, ahora estoy muerto? (said in a mocking tone)
‘Now you are really making it even better, so am I dead now?’
Ca/Jo: (laugh)
C: Bueno, pues eso, que reconstruían cómo era y era súper guapo, con los ojos marrones, enormes, y me recordó mucho a ti.
‘Well, just that, they recreated how he looked like and he was super handsome, with huge brown eyes and he reminded
me lots of you.’
!
J: Pues haber empezado por ahí!
‘You should have started there!’
C: Claro, si por ahí iba yo pero como algunos me han interrumpido. . .
‘Sure, that's where I was heading but some people have interrupted me. . .’
Ca: Anda, cariño, no te mosquees.
‘Come on, sweetie, don’t get pissed off ’
C: pero si no me mosqueo, como sois de malos (laughs)
‘I’m not pissed off; you really are a wicked bunch’
(All laugh and the conversation continues)

Example (11) above is particularly interesting since it involves four interlocutors and face-threat becomes ‘‘more
public’’ than in those cases where only the speaker and hearer are involved. In this particular case, the presence of Jo
makes it even more embarrassing to J, since it is the first time they meet and both want to make a good impression.
Furthermore, it could be argued that speaker C's object of comparison is rather infelicitous not only to the targeted hearer
(as his slightly offended mode shows) but also to the other two participants, who laugh at it and even make their own
comments. In other words, the complimenter's infelicitous comparison (albeit unintended) poses a face-threat not only to
the complimentee but also to the complimenter herself, whose second conversational turn shows her attempt to repair
face (both hers and the hearer's). Her embarrassment, together with all her false starts and pauses, shows that her face
has also been damaged in the misunderstanding.
In all the examples previously analyzed, the complimented entity is either a personal achievement (example 6) or the
interlocutor's appearance (examples 7--11). The speaker could have opted for a more direct -- formulaic -- compliment but
chose instead a riskier option. Acting this way, the speaker tries to avoid making too direct, face-threatening remarks that
could place the hearer or complementee in an embarrassing situation. In fact, at least in the corpus used for the present
analysis, there are no examples of implicit compliments used when complimenting new possessions or goods, where
explicitness seems to be preferred both in English and Spanish. As already pointed out, this might be due to the fact that
showing appreciation for an object might be considered much less face-threatening than when evaluating someone's
qualities, achievements or personal appearance. This is very much in line with Kerbrat-Orecchioni (1987) and Herbert's
(1990) views. As pointed out by Bruti (2006:195--196),
[Implicit compliments] certainly reduce the potential threat to the addressee's negative face that often emerges in
the form of embarrassment in front of overt compliments. In fact, on such occasions the complimentee tends to
respond so as to downplay the expressed praise. On the contrary, with implicit compliments the complimentee finds
it less difficult to accept the content of the assertion made by the speaker.
Following Boyle (2000), the second most common way to pay an implicit compliment in English is to refer to something the
interlocutor has done and which they are proud of, as illustrated by the following examples:

Example 12 (Taken from Boyle, 2000):

‘‘You’ve worked with Elizabeth Taylor!’’

Example 13 (Taken from Bruti, 2006):

D: Well, I may know nothing of Dow Jones, but I do know something about kisses.
L: You could lecture on that at Vassar.

Quite surprisingly, this pattern is absent in our corpus. Moreover, compliments aimed at evaluating the interlocutor's
personal characteristics and/or deeds are also practically non-existent in our corpus. In fact, there are just a couple of cases
where the addressee is complimented on personal characteristics (e.g. her kindness), as illustrated by example (14)17:

17
There is also one example where the speaker compliments the addressee on a personal achievement. See example (6).
988 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996

.Example 14:
[Field: C is driving her friend E. However, C gets lost and this makes E take longer than if she had opted for public transport
(her first idea before C offered to give her a lift). C's first conversational turn is an apology to which E graciously responds
with an implicit compliment about the pleasure of her friend's company.]

C: Al final, vas a tardar más que si hubieras ido en transporte público.

‘In the end, it's going to take you longer than if you’d gone by public transport.’
E: Sí, pero no hubiera sido tan agradable.
‘Yes, but it wouldn’t have been so nice.’
C: Uy, vaya, gracias.
‘Oh dear, thanks.’
E: Es que es verdad.
‘It is but true.’

In example (14) above, E's response is pragmatically very appropriate since it succeeds in restoring C's face loss both
by accepting the apology and by sincerely showing that she likes her friend's company, even if this entails an unexpected
loss of time. On the other hand C's response (‘oh, dear, thanks’) clearly shows that she has correctly interpreted the
previous comment as a compliment and possibly as a restoration of face.
The limited size of the corpus makes it impossible to argue whether the absence of this type of compliments (i.e. on
personal abilities) is due to a different pragmatic orientation or whether it is simply coincidental. However, previous
research on formulaic compliments showed that Spanish speakers tended to compliment more on aspects such as new
items or appearance than skills or abilities (Maíz-Arévalo, 2010).
As already mentioned, Boyle (2000) identifies two major patterns in English:

(i) Comparison with a third entity

(ii) Referring to something the interlocutor has done and which s/he is proud of.

As we have seen, the data show that the first pattern is also the most frequent in Spanish (54% of the total) whereas
the second one is absent. This absence, however, can be explained by the emergence of other patterns in the analysis
of the present corpus. In fact, it is possible to identify three more patterns together with comparison with a third entity.
Thus, the data reveal that Spanish provides the speaker with the following choices when paying an implicit compliment:

(i) Comparing the addressee with a third entity (54% of the cases)
(ii) Complimenting a third entity connected to the addressee (27% of the examples)
(iii) Comparison with omission of the second term of the comparative (11.5% of the cases)
(iv) Asking the interlocutor an apparently irrelevant question (7.5% of the cases)

The second most common pattern is based on complimenting a third entity who may be present or not at the moment of
paying the compliment. Both interlocutors know that this third entity is connected to the addressee, to whom the
compliment is really targeted. This pattern is also present in English, as Holmes indicates when discussing implicit
compliments (1995:117):

R: What a polite child!

C: Thank you. We do our best.

In this example, C correctly interprets R's exclamation as a compliment to herself since she is the child's mother and is
considered to be doing a good job in his/her education. In the Spanish corpus, we find examples where the addressee's
appearance is complimented by means of what we could define as a ‘‘vicarious comparison’’. In other words, someone
closely related to the addressee is complimented on appearance, implying that the addressee is the origin of that beauty.
This is the case of examples (15) and (16) below:

Example 15:

[Field: M and Cn work together but there is not any type of personal relationship behind work. On that particular occasion,
M, who is on holiday, has gone to work for a visit and brought her little niece with her. The exchange is part of the greeting. /
Tenor: Cn (male-in his late 30s) and M (female in her late 30s)]
 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996 989

Cn: Vaya ojazos tiene esta cría. Desde luego, tiene a quien parecerse.
‘This kid's got really beautiful eyes. Sure, she has someone to look like.’
M: Uy, gracias.
‘Oh, thanks.’
Example 16:
[Field: A and C work together but there is not any type of personal relationship behind work. C and A have bumped into
each other outside work and they greet and chitchat for some minutes. C is accompanied by her stepdaughter, whom
A mistakes for her own daughter. / Tenor: A (male-in his early 40s) and C (female in her late 30s)]
A: Qué ojos más bonitos tiene la niñ a. Tiene a quien salirle.
‘What lovely eyes the kid has. She takes after someone’
C: Gracias por la parte que me toca pero no soy su madre.
‘Thank you for what could be my doing but I am not her mother’
A: Vaya, qué metedura de pata, no tenía ni idea. Lo siento.
‘Oh dear, how embarrassing! I had no idea. I’m sorry.’
C: No hombre, tranquilo, si no sabías nada. Además, te agradezco el cumplido de todas formas.
‘No, come on, don’t worry, you knew nothing. Besides, thanks for the compliment anyway.’
A: Es que tú también tienes unos ojos preciosos.
‘Yes, you also have such beautiful eyes.’
C: (smiles)
In example (16), A's wrong assumption -- i.e. that C's accompanying kid is her daughter -- leads to what he considers a
loss of face, quickly restored by C. We also know that she has interpreted his comment correctly as a compliment since
she explicitly thanks him for it in the same conversational turn. Once he feels his face has been restored, A reformulates
the compliment in a more explicit (formulaic) way, maybe because he considers their relationship has advanced a step
forward and is not that distant any longer since C is sharing with him information about her personal life. This renegotiation
of distance makes A feel entitled to pay more explicit, direct compliments to C.
Besides complimenting appearance, this pattern is also used when complimenting someone in general; that is, it is not
totally clear what is being complimented -- i.e. appearance, skills, personality or the three of them simultaneously as
illustrated by example (2) (repeated here as (17) for the sake of clarity):
Example 17:
[Field: Two work colleagues are saying farewell after a celebratory lunch together. C has told K that she is engaged.//
Tenor: K (male, around 45) and C (female, 38) are not close friends but their working relationship is developing into
friendship. However, they still have not met outside work.]
K: Oye, lo dicho, que me alegro mucho de lo de tu boda. Tu prometido es un hombre muy afortunado.
‘Listen, let me just tell you again, I’m really happy about your wedding. Your fiancé is a very lucky man.’
C: Hombre, muchas gracias.
‘Man, thanks a lot.’
The third pattern revealed by the data consists of omitting the second term of the comparative as already illustrated by
example (14) above. In these comparisons, the second term is elided and left for the hearer to fill in. The implicature is thus
triggered by the use of lexical items like ‘‘tan’’ (‘so’ or ‘as’), ‘‘tanto’’ (‘so much’), etc. as is further illustrated by examples
(18)--(21) below18:
Example 18:
[Field: Speaker Ca is meeting his in-laws and he obviously wants to make a good first impression without sounding like a
flatterer. // Tenor: Ca (male, 38) and his fiancée C (female, 38), who only takes part in the exchange as a listener, her
parents M (female, 65) and P (male, 64)]
C: Encantado de conoceros. Ahora sé de dónde saca Cr tanta belleza.
‘Pleased to meet you. Now I know where Cr's beauty comes from.’
M: ay, qué majo tu novio. Eso dicen, que C se parece mucho a mí.
‘Aw, how nice is your boyfriend. So they say that C looks a lot like me.’

18
In Relevance Theory, this process is known as free enrichment. In Yus’ words (2011:145), ‘‘it takes place when the utterance demands from
the hearer the ‘inferential filling’ of some elided part.’’
990 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996
? ?
P [addressing his wife, M]: oye, y a mí qué? A mí no se parece?
‘Listen what about me? Doesn’t she look like me?’
(All of them laugh and change topic)

Quite curiously, this type of compliments is also reported in English in exactly the same kind of situations (i.e. meeting
the in-laws), as illustrated by example (19) (taken from Bruti, 2006):

Example 19:
Mrs. Shanahan: Nice to meet you, Hal.
Hal: The pleasure is mine, Mrs. Shanahan. Wow. I can see where Rosemary gets her figure.

Example 20:
[Field: E and C have just come back from a date. E has given C a lift home and is saying farewell. // Tenor: E (male, 30) and
C (female, 33). // C's response (‘‘gracias’’) shows that she has taken E's previous comment as a compliment to herself.]

E: Me lo he pasado genial. Hacía mucho que no me divertía tanto.

‘I had a great time. It was ages since I didn’t have so much fun’
C: Gracias
‘Thanks’

Example 21:
[Field: E and C work together and share the same office, where E has recently moved. E was looking for some paper in C's
drawer. // Tenor: E (female, 33) and C (female, 38). Their relationship is developing forward towards friendship although
they still meet mostly at work.]

E: Qué barbaridad, ya me gustaría a mi ser tan organizada. . .

‘Oh dear, I’d love to be so organized. . .’
C: Qué va! pero si no soy nada organizada. . .
!
‘No way! I’m far from organized. . .’
E: Que sí, además, lo decía como un cumplido.
‘Sure you are, besides, I meant it as a compliment.’
C: Ya, bueno, gracias.
‘I know, well, thanks.’

Looking at these examples, it could be argued that this pattern is pragmatically less risky than simply establishing a
comparison with a third entity since there is no room for ambiguity. As we have already seen, for a comparison with a third
entity to be felicitous it is necessary for both the speaker and the hearer to share the same admiration for that third entity. In
this second pattern, however, the implicature does not rely on shared background knowledge but simply on a lexico-
grammatical structure (‘‘tan(to). . . como’’ ‘as (much). . .as’) where the second term of the comparison is easily retrieved by
the hearer through the process of free enrichment (see note 18). This is especially clear in example (21), where E's implicit
compliment (‘‘you are very organized’’) is correctly interpreted as such by C, who responds with the typical pattern of
rejection (‘‘no way!’’) (see Maíz-Arévalo, in press). E then moves on to repeat the compliment in a more explicit way,
adding a meta-pragmatic comment (‘‘Lo decía como un cumplido’’ ‘I meant it as a compliment’) which is considered
obvious by C (‘‘Ya’’ ‘I know’). The compliment is finally accepted and both interlocutors change topic.
Occasionally, however, face-threat cannot be avoided even if the compliment is implicit, as demonstrated by the
following example:
Example 22:
[Field: IN has just arrived at his mother's place and is surprised to find C and Cr. The exchange takes place as part of his
greeting to them. Although only two interlocutors speak in this extract, there were six people present (see tenor) // Tenor:
IN (male-35) and Cr (male-38) are brothers, although their relationship is not particularly good. Cr is about to marry C
(female-38). She knows IN but they are still not close since they have only met occasionally. Apart from them, the other
interlocutors present are IN and Cr's mother (in her sixties) and IN's wife (female-in her late thirties). However, at the
moment of the exchange between IN and C, none of them seemed to be paying any attention to their ‘‘repartee’’. IN refers
to M (female, in her late twenties), who is his mother's personal caretaker and, therefore, not properly part of the family but
an employee. Besides, M is not particularly beautiful but overweight]
 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996 991

!
IN: Anda, qué sorpresa! Si estáis aquí. Ya decía yo que M. no podía estar tan buena.
‘Wow, what a surprise! It is you here. I wondered how M. could look so hot.’
C: [clearly embarrassed] ah, sí, bueno. . .
‘Ah, yes, well. . .’
? ?
IN: bueno, y qué tal? todo bien? (changes topic)
‘Well, what's up? I everything ok?’

Example (22) also illustrates a recurrent pattern in Spanish where greetings are usually accompanied by a compliment
(Maíz-Arévalo, 2010). In this case, the interlocutor (IN) opts for an implicit compliment based on the omission of the second
term of the comparison (‘‘estar tan buena [como tú] ’’, ‘to be as hot [as you]’). The addressee (C) is left to infer the second part
‘‘como tú’’ (‘like you’). Her vague response, however, shows that she has correctly inferred the compliment but finds it
embarrassing. The infelicity of the situation is also correctly ascertained by IN, who quickly chooses to change the topic.
There are two main reasons why this compliment leads to socio-pragmatic failure19 (Thomas, 1983). First, although it is a
compliment on C's appearance (concretely on her beauty), the expression chosen (‘‘tan buena’’) has a sexual tint in
Peninsular Spanish, inappropriate between future relatives. In fact, the interlocutor could have chosen more neutral
expressions like ‘‘tan guapa’’ (‘so pretty’). Related to this is the fact that the interlocutors have known each other for some
months but their relationship is still not too close to pay such ‘‘intimate’’ compliments. To his credit, speaker IN immediately
realizes his compliment has gone too far and chooses to restore the breakdown (and both interlocutors’ face-loss) by
changing the topic and initiating a new ‘greeting’ sequence, more in accordance with the kind of relationship both
interlocutors still share.
Finally, the fourth pattern revealed by the data is based on using a totally different strategy. In these cases, speakers
choose interrogative sentences to ask the interlocutor what seems like an apparently irrelevant20 question. Pragmatically,
the choice of this linguistic realization -- i.e. an interrogative clause -- entails an immediate involvement of the addressee in the
exchange. In other words, questions demand an answer and therefore a participation in the conversation, which makes them
particularly suitable to involve the addressee, who is hence required to provide an answer. Since one of the primary functions
of compliments (both implicit and explicit) is to create rapport with the addressee, it is not hard to understand why interrogative
clauses might be used for this purpose. This pattern is illustrated by examples (23) and (24) below:

Example 23:

[Field: L and M work together but their relationship is purely professional. This is part of a greeting between both. // Tenor: L
(male, in his late 50 s) and M (female, in her late thirties).]
?
L: Vaya, dónde es el pase de modelos?
‘Wow, where is the catwalk?
M: uy, vaya, gracias por el cumplido
‘Oh, dear, well, thanks for the compliment’

Example 24:

[Field: J and M work together but their relationship is purely professional. This is part of an early morning greeting between
both. // Tenor: J (male, in his mid 40 s) and M (female, in her late thirties).]
?
J: Uy, qué se celebra?
‘Oh, what's the occasion?
?
M: Cómo que qué se celebra?
‘What do you mean what's the occasion?’
J: Mujer, lo decía porque como has venido tan elegante. . .
‘Well, I said so because you look very elegant. . .’
M: Ah, vaya, gracias. Estoy dormida. . .
‘Oh, dear, thanks. I am still sleepy. . .’
J: (laughs) Bueno, a estas horas es normal [change of topic]
‘Well, it's normal at this time of day’

19
According to Thomas (1983), socio-pragmatic failures are those where the interlocutor performs a speech act in the wrong social context.
20
Following Sperber and Wilson (1986, 1995).
992 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996

As in many other cases, these compliments always co-occur with greetings (at least in the present corpus). Since greetings
in Spanish are usually accompanied by compliments, these apparently irrelevant questions are hence likely to be interpreted
by the addressee as compliments accompanying the greeting, as is the case of example (23). Here, M correctly infers
that the implicature is that she looks especially elegant and therefore, should know where the catwalk is. However, it
could be argued that this type of compliment also seems to be slightly more difficult to process for interlocutors. This
could also account for the fact that they are less frequently used (only 7.5% of the total). In fact, this is clearly the case in
example (24) where the addressee admits that she is still cognitively impaired and does not understand the illocutionary
force of J's question. This forces him to reformulate the compliment in a more explicit, non ambiguous way which is
correctly interpreted by M.
Other examples reveal that speakers are well aware that this kind of compliments can miss their point, leading them to
reformulate their own compliment in the same conversational turn, even before the addressee is given a chance to
respond. This is illustrated by example (25) below, where the speaker (Jp) uses different realizations. Thus, he first
employs an apparently irrelevant question followed by comparison with a third entity before rounding all off with an explicit
compliment. This is all carried out in the same conversational turn, before even giving C the chance of showing any
inferences she may have made. The reason why speaker Jp acts like this could be that he feels that C might lack all the
background information -- i.e. she may not have seen that film or know this actress -- which compels him to become
increasingly more explicit.

Example 25:

[Field: Jp and C work together but their relationship is slightly becoming more personal than purely professional. This is
part of an early morning greeting between both after C has been to the hairdresser's and changed her look. // Tenor: Jp
(male, in his late 30 s) and C (female, in her late thirties).]
?
Jp: Anda has visto La puta y la ballena? pues estás clavada a Aitana Sánchez Gijón. Estás mejor así, mucho más guapa.
‘Oh, have you seen ‘‘The whore and the whale’’? Well, you look just like Aitana Sánchez Gijón. You look better this way,
much prettier.’
C: Uy, vaya, gracias.
‘Oh, well, thanks.’

5. Speakers’ pragmatic motivations

In general terms, compliments (both implicit and explicit) have long been regarded as one of the clearest examples of
positive politeness, since they pay attention to the hearer's need to be approved of and liked by the speaker(s). This view
has been shared by numerous authors (Wolfson, 1981, 1983; Holmes, 1986, 1987; Herbert, 1989; Lewandowska-
Tomaszczyk, 1989; Johnson and Duane, 1992). In this sense, compliments’ primary function is to create and/or maintain
rapport between the interlocutors’ (Wolfson, 1983:86).
However, it is also well known that compliments, far from their apparent innocent aspect, involve many different
variables that can turn them into a minefield for interlocutors, especially if they belong to different cultures (Holmes and
Brown, 1987). In fact, the speaker who chooses to pay a compliment has to take into account many aspects such as: the
social relationship between the speaker and the addressee, what exactly can be complimented and what should not, the
point of the conversational exchange, the gender and age of the interlocutors and so on.
Following Brown and Levinson (1987), it can therefore be argued that compliments are highly face-threating acts since
they can threaten both interlocutors’ face if the compliment is inappropriate -- e.g. what is complimented is too personal as
in example (26 below):

Example 26:

[Field: A is part of the international audience to a conference. B has just delivered her talk and is approached by A, who
wants to congratulate her. Tenor: A (female, in her early 30s) and B (female, in her mid-thirties).]

A: Very interesting talk, I quite enjoyed it.

B: Thank you, I’m glad you liked it. (longer pause)
A: That's a lovely ring (touching it)
B: huh, hum, thank you (visibly embarrassed)
(Maíz-Arévalo, 2010:176)
 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996 993

In this example, the social relationship between the interlocutors is clearly distant since they have never met before and
their only link is the fact that they are both colleagues in the same field and attending the same conference together. In this
context, an explicit compliment on the lecture is regarded as appropriate and the addressee reacts positively, thanking the
speaker. However, this is not the case of the second compliment which is clearly seen as inappropriate by B, who deems it
too intimate and personal and responds to it visibly embarrassed. In this case, it can be argued that both the speaker's and
the addressee's face are being threatened. Indeed, the addressee's face feels threatened because the speaker assumes
a closer relationship than they really have and enters personal fields (she goes as far as establishing physical contact).
Likewise, the speaker's face feels threatened because her compliment has been considered inappropriate and ‘crossing
the line’.
Avoiding face-threat might thus be one of the major reasons why a speaker opts for paying an implicit compliment,
especially in those contexts where the relationship between the interlocutors is relatively distant.21 Interviews with some of
the participants indeed support this idea, as shown by their own words:

Interviewee 1
La verdad es que no lo había pensado, pero prácticamente acabamos de conocernos. Supongo que es por eso.
‘The truth is that I hadn’t given it a thought, but we have practically just met. I guess that's why.’

Interviewee 2
Uy, pues no sé, me salió así. Tampoco la conozco tanto.
‘Well, I don’t know, it came out this way. It's not that I know her that well.’

In fact, the data reveal that most of the analyzed exchanges took place between work colleagues whose relationship
was either purely professional or starting to develop into a more personal one. There are other cases in the corpus,
however, where the participants involved share a closer relationship. In this case, distance cannot be the variable that
determines why the speaker chooses to be conventionally implicit instead of opting for the most common formulaic
compliment. However, according to Lewandowska-Tomaszczyk (1989) indirectness might also be preferable as a more
elaborated option where the speaker's extra effort in paying a compliment also shows their personal involvement and
sincerity. In some of the interviewees’ words:

Interviewee 3
Bueno, ahora que lo dices, no sé muy bien por qué lo dije de esa forma, supongo que, bueno, que quería quedar bien,
caerles bien, pero como todavía no los conozco mucho, no me atrevía a soltarlo así, directamente y que sonara a peloteo
barato.
‘Well, now you mention it, I really don’t know why I spoke that way, I guess that, well, I wanted to make a good impression, to
be liked, but since I still don’t know them that much, I didn’t dare put it like that, directly and that it sounded like cheap flattery.’

Interviewee 4
Era la primera vez que me los presentaba y quería decirles cosas agradables pero sin sonar demasiado forzado, falso.
‘It was the first time I met them and I wanted to say nice things to them but without sounding too forced, fake.’

Interviewee 5
Hay gente que no para de hacerte cumplidos cuando te conoce por primera vez, como para congraciarse contigo. Qué
guapa esto, qué bonita tu falda, al final, suena falso, como si fuera peloteo. Como no me gusta cuando me lo hacen,
intento no hacerlo yo también, aunque supongo que a veces nos sale sin pensar.
‘There are people who can’t stop paying you compliments when they meet you, as if they were trying to ingratiate
themselves with you. How pretty this, what a lovely skirt, in the end, it sounds fake, just flattery. Since I don’t like it when it's
done to me, I try not to do so myself, although I guess it sometimes comes out without our thinking.’

According to Brown and Levinson (1987), one of the ways of showing politeness is appealing to or claiming common
ground. In those cases where the relationship between the participants is becoming increasingly closer or where it is
already close, the choice of a conventionally implicit compliment where common knowledge is assumed can be extremely
appropriate from the pragmatic point of view. It reinforces common ground whilst at the same time sounding more sincere.
This is what interviewee 6 pointed out when she recalled her own words (referring to example 6):

21
In this sense, conventionally implicit compliments can be paralleled with conventionally indirect requests, where indirectness is used to
redress face-threat (Brown and Levinson, 1987).
994 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996

Interviewee 6
Bueno, cada vez nos llevamos mejor, la verdad es que yo diría que casi se nos puede llamar amigas, o por lo menos, ya
no somos solamente compañ eras de trabajo porque hablamos de muchas cosas, de otras cosas que no son solo del
trabajo. Lo hizo muy bien, sinceramente, y quería que lo supiera, que de verdad pensaba eso. Compararla con J, bueno,
creo que es el mejor cumplido, porque las dos la admiramos muchísimo.
‘Well, we get on better and better, the truth is that I’d say we are practically friends now, or at least, we aren’t just work
colleagues because we talk about many things, other things which are not related to work. She did very well, honestly, and
I wanted her to know, that I really thought so. Comparing her with J, well, I think it is the best compliment, because both of
us admire her so much’

To recap, this choice in the system of compliments is there because it serves to fulfill a certain range of functions,
among the most important being: avoiding face-threat (especially when distance is still relatively high between
interlocutors), wishing to sound sincere and/or building rapport. Thus, we can further elaborate the system of compliments
[(Fig._2)TD$IG]
(see Fig. 2) so as to include implicit compliments as follows.

BASED-ON- comparison-with-third-entity
based-on-comparison
COMPARISON-TYPE omission-of-comparative-second-term

based-on-implicit-relationship-between-h-and-a-third-entity
IMPLICIT_COMPLIMENT-
implicit_compliment
TYPE irrelevant-question

questions like "what's the occasion?"

reference-to-speaker's-feelings-and-emotions

Fig. 2. System of implicit compliments.22

6. Conclusions

Drawing on Boyle's seminal study on implicit compliments, the present paper has aimed to redress the balance in favor
of implicit compliments in Peninsular Spanish. Far from being deviant, implicit compliments open up a system of choices
(forms) which serve to fulfill a range of functions such as the wish to show sincerity and build up rapport between
interlocutors. Factors like these together with others like the degree of distance between speaker and hearer have a
crucial say in the speaker's choice for this much rarer speech act, even if this choice can occasionally turn wrong, as some
of the analyzed examples have shown.
Together with ascertaining how interlocutors prefer implicature to avoid face-threat (either theirs or their
complimentees’), findings also reveal that some implicit compliments may not be that implicit after all, but repeat the
same patterns -- even their linguistic realization -- both in English and in Spanish. This leads us to argue for a cline of
explicitness when dealing with compliments, ranging from the most formulaic types (e.g. what lovely eyes!) to those based
on implicatures more difficult to decipher.23 In between, however, we find what can be termed conventionally implicit
compliments with examples like ‘‘he's a lucky fellow’’ or ‘‘I see where she gets her beauty from’’ that are paralleled in both
languages and in exactly the same situations, which supports the claim that implicit compliments are far from deviant
cases and obey complex pragmatic motivations.
While the pragmatic motivations seem to be the same in English and Spanish, the present analysis reveals a significant
difference between both languages. With regard to what is complimented, Spanish interlocutors seem to compliment
more on appearances than abilities both explicitly (see Maíz-Arévalo, 2010) and implicitly. On the other hand, one of the
two major patterns of implicit compliments in English ‘‘pivots around the addressee's achievement’’ (Bruti, 2006:188).
Two final limitations should be admitted, though. On the one hand, the relatively small size of the corpus cautions us to
simply comment on the tendencies observed rather than rushing into simplistic generalizations. On the other, it is beyond
the scope of this paper to analyze deeply other important variables like gender, age or the status of the participants. Future
research intends to delve into these variables and ascertain the role they play in the production of conventionally implicit
compliments.

22
Fig. 2 has been generated by means of Corpus Tool, designed by Mick O’Donnell (2007) and available at: http://www.wagsoft.com/
CorpusTool/.
23
See example (3). It must be admitted, however, that purely implicit compliments are extremely challenging since the receptors (let alone the
researcher) might not even be aware of their illocutionary force. On the other hand, no examples have been found in the present corpus.
 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996 995

Acknowledgments

I would like to express my gratitude to all the participants who have kindly granted me permission to use their examples,
especially to those who agreed to be interviewed and whose intuitions made my arguments more believable. My special
thanks go to my future husband for his infinite patience, support and encouragement. Last but not least, I would like to
thank Professor Culpeper, Professor Arús Hita and the two anonymous reviewers, whose stimulating suggestions and
comments helped me widen the scope of this paper and see the analysis with more thoughtful eyes. Needless to say, all
remaining errors are my own.

References

Boyle, Ronald, 2000. ‘‘You’ve worked with Elizabeth Taylor!’’: phatic functions and implicit compliments Applied Linguistics 21 (1), 26--46.
Brown, Penelope, Levinson, Stephen, 1987. Politeness. Cambridge University Press, Cambridge.
Bruti, Silvia, 2006. Cross-cultural pragmatics: the translation of implicit compliments in subtitles. The Journal of Specialised Translation 6, 185--
197.
Chen, Rong, 1993. Responding to compliments. A contrastive study of politeness strategies between American English and Chinese speakers.
Journal of Pragmatics 20 (1), 49--75.
Golato, Andrea, 2002. German compliment responses. Journal of Pragmatics 32, 29--54.
Golato, Andrea, 2003. Studying compliment responses: a comparison of DCTs and recordings of naturally occurring talk. Applied Linguistics 24
(1), 90--121.
Golato, Andrea, 2005. Compliments and Compliment Responses. John Benjamins, Amsterdam.
Halliday, Michael A.K., Hassan, Ruqaiya, 1976. Cohesion in English. Pearson Education Ltd, Malaysia.
Herbert, Robert K., 1989. The ethnography of English compliments and compliment responses: a contrastive sketch. In: Oleksy, W. (Ed.),
Contrastive Pragmatics. John Benjamins, Amsterdam, pp. 3--35.
Herbert, Robert K., 1990. Sex-based differences in compliment behavior. Language in Society 19, 201--224.
Herbert, Robert K., 1991. The sociology of compliment work: an ethnocontrastive study of Polish and English compliments. Multilingua 10 (4),
381--402.
Herbert, Robert K., Straight, H.S., 1989. Compliment-rejection versus compliment-avoidance: listener-based versus speaker-based pragmatic
strategies. Language and Communication 9 (1), 35--47.
Holmes, Janet, 1986. Compliments and compliment responses in New Zealand English. Anthopological Linguistics 28 (4), 485--508.
Holmes, Janet, 1995. Women, Men and Politeness. Longman, London and New York.
Holmes, Janet, 1998. Paying compliments: A sex-preferential positive politeness strategy. Journal of Pragmatics 12, 445--465.
Holmes, Janet, Brown, Dianne, 1987. Teachers and students learning about compliments. TESOL Quarterly 21 (3), 523--546.
Jaworski, Adam, 1995. ‘‘This is not an empty compliment!’’ Polish compliments and the expression of solidarity International Journal of Applied
Linguistics 5 (1), 63--94.
Johnson, Donna, Duane, Roen H., 1992. Complimenting and involvement in peer reviews: Gender variation. Language in society 21, 27--57.
Kerbrat-Orecchioni, Catherine, 1987. La description des échanges en analyse conversationalle: l’exemple du compliment. DRLAV Revue de
linguistique 36, 1--53.
Knapp, Mark L., Hopper, Robert, Bell, Robert A., 1984. Compliments: a descriptive taxonomy. Journal of Communication 34 (4), 12--32.
Lewandowska-Tomaszczyk, Barbara, 1989. Praising and complimenting. In: Oleksy, W. (Ed.), Contrastive Pragmatics. John Benjamins,
Amsterdam, pp. 73--100.
Liu, David, 1995. Sociocultural transfer and its effect on second language speakers’ communication. International Journal of Intercultural
Relations 19 (2), 253--265.
Lorenzo-Dus, Nuria, 2001. Compliment responses among British and Spanish university students: a contrastive study. Journal of Pragmatics 33,
107--127.
Maíz-Arévalo, Carmen, 2010. Intercultural Pragmatics: a contrastive analysis of compliments in English and Spanish. In: Gómez, Blanco, Luisa,
Ma, Marín Arrese, J. (Eds.), Discourse and Communication: Cognitive and Functional Perspectives. Dykinson & Universidad Rey Juan Carlos,
Madrid, pp. 175--208.
Maíz-Arévalo, Carmen, in press. At a loss for words or how to respond to compliments: a contrastive analysis of compliment response in English
and Spanish. In: Padilla Cruz, Manuel, et al. (Eds.), (Im)politeness and Interpersonal Communication. Cambridge Scholars, Sheffield.
Maíz-Arévalo, Carmen, García-Gómez, Antonio. ‘You look terrific!’: social evaluation and relationships in online compliments, forthcoming.
Moore, Zena, 1996. Teaching culture: A study of piropos. Hispania 79 (1), 113--120.
Plasencia, Ma Elena, Yépez Lasso, Mariana, 1999. Compliments in Ecuadorian Spanish. Lengua 9, 83--121.
Pomerantz, Anita, 1978. Compliment responses. Notes on the co-operation of multiple constraints. In: Schenkein, J. (Ed.), Studies in the
Organization of Conversational Interaction. Academic Press, London, pp. 79--112.
Saito, Hidetoshi, Beeken, Masako, 1997. An approach to instruction of pragmatic aspects: implications of pragmatic transfer by American learners
of Japanese. Modern Language Journal 81 (3), 363--377.
Searle, John, 1969. Speech Acts: An essay in the philosophy of language. Cambridge University Press, Cambrigde.
Sperber, Dan, Wilson, Deirdre, 1986 (1995). Relevance. Communication and Cognition. Blackwell, London.
Thomas, Jennifer, 1983. Cross-cultural pragmatic failure. Applied Linguistics 4, 91--112.
Valdés, Guadalupe, Pino, Cecilia, 1981. Muy a tus órdenes: compliment responses among Mexican-American bilinguals. Language in Society 10
(1), 53--72.
Wolfson, Nessa, 1981. Compliments in cross-cultural perspective. TESOL Quarterly 15 (2), 117--124.
996 C. Maíz-Arévalo / Journal of Pragmatics 44 (2012) 980--996

Wolfson, Nessa, 1983. An empirically based analysis of complimenting in American English. In: Wolfson, Nessa, Judd, Elliot (Eds.),
Sociolinguistics and Language Acquisition. Newbury House Publishers, Tokyo, pp. 82--95.
Wolfson, Nessa, Manes, Joanne, 1980. The compliment as a social strategy. Papers in Linguistics 13 (3), 391--410.
Ylänne-McEwen, Virpi, 1993. Complimenting behaviour: a cross-cultural investigation. Journal of Multilingual and Multicultural Development 14
(6), 499--508.
Yuan, Yi, 2002. Compliments and compliment responses in Kunming Chinese. Pragmatics 12 (2), 183--226.
Yus, Francisco, 2011. Cyberpragmatics. John Benjamins, Amsterdam.

Carmen Maíz-Arévalo obtained her PhD in English Linguistics in 2001, being an English teacher since 1995. Currently Dr. Maiz-Arevalo holds
the position of full time lecturer at the Universidad Complutense, where she is teaching Pragmatics, Semantics and English. Her fields of interest
are mainly speech act theory, politeness, gender and language and applied linguistics. She has published several articles on these issues and
taken part in numerous congresses, both national and international. Besides her research and teaching activities, Carmen Maiz-Arevalo is also
the secretary of the Revista de Estudios Ingleses, published yearly by the Universidad Complutense de Madrid.