Interview

Rachel Giora *
Edited by Francesca Ervas
[email protected]
Rachel Giora is Professor of Linguistics at Tel Aviv University. Her work has
been devoted to exploring the ways salient meanings of words shape how we
think and speak. Giora analyzes meaning salience in both figurative and literal
language. The main question around this general topic is the way in which,
while words have multiple meanings, some meanings are more accessible than
others. Given the notion of graded saliency, access of information stored in the
mental lexicon is therefore ordered: more salient meanings are accessed before
less salient meanings. Degree of salience is determined by factors such as
frequency of use, experiential familiarity, conventionality, prototypicality, etc.
Giora argues that both literal and non-literal meanings that are salient are
cognitively prominent salient meanings and therefore they play a very
important role in the comprehension and production of language. Her work
focuses on the psycholinguistics of figurative language (irony, jokes, and
metaphor), context effects, optimal innovations and aesthetic pleasure,
discourse negation, context and degree of salience. One of her most popular
books is On Our Mind: Salience, Context, and Figurative Language,
published by Oxford University Press in 2003.
1. In On our mind (Giora, 2003), you analyzed a variety of figurative
language cases, such as metaphors, idioms, and jokes, paying
attention also to the role of context. To what extent does context
influence figurative language comprehension? Are there contexts
which favor non-literal interpretation?
No theory dismisses the role of contextual information in utterance
interpretation. The debate, however, revolves around the timing of its effects:

* Department of Linguistics, Tel Aviv University, Israel.

Department of Education, Psychology, Philosophy, University of Cagliari, Italy.

Humana.Mente Journal of Philosophical Studies, 2012, Vol. 23, 189196

190

Humana.Mente Issue 23 December 2012

Are context effects initial, relevant to early comprehension processes, or are

they late, affecting only the products of early processing? On one view - the
direct access and constraint-satisfaction models - if contextual information is
strongly supportive, comprehenders can immediately and directly derive the
appropriate interpretation (e.g., Campbell & Katz, 2012; Colston, 2000;
Gibbs, 1986, 1994, 2002; Pexman, Ferretti & Katz, 2000; Katz & Pexman,
1997; Utsumi, 2000). On another, contextual information has no effect on
initial processing. Instead, an obligatory literal stage is involved initially, even
when contextually inappropriate (e.g., Grice, 1975; Searle, 1979). On the
Graded Salience Hypothesis (Giora, 1997, 1999, 2003; Giora, et al. 2007),
even a strong context cannot block salient meanings, and hence salience-based
interpretations interpretations based on the salient meanings of the utterance
component - regardless of (non)literality.
But what is a strong context? Or, put in your words, is there a specific kind
of context that favors non-literal interpretation? First, as you have said, there is
no just one kind of non-literal language. Its not just the difference between
metaphors, sarcastic ironies, proverbs, jokes, etc. that matters. According to
the Graded Salience Hypothesis, the psychologically relevant distinction
relates to degree of salience, regardless of degree of (non)literality. Given that
salient (coded and prominent) meanings, whether literal or non-literal, will
always be activated initially, regardless of contextual information, the question
as to what context favors which interpretation is relevant only to meanings and
interpretations low on salience. It is no wonder then that research focusing on
contextual effects has dealt primarily with nonconventional expressions and
utterances. Sarcasm or verbal irony has attracted most of the attention.
The question as to whether there is a specific type of context that invites a
sarcastic interpretation has been treated by proponents of the direct access
view and the constraints-satisfaction model. According to Katz and his
colleagues, a context rich in sarcastic cues should facilitate sarcasm initially.
Among these cues are speakers occupations, failed expectation, pragmatic
insincerity, negative tension, presence of a victim, to name a few. While none
of them is necessary, a context employing multiple such cues will favor a
sarcastic interpretation (Campbell & Katz, 2012). According to Gibbs
(2002), it is a protagonists failed expectation that induces an expectation for a
sarcastic utterance and consequently - a sarcastic interpretation.
In a number of studies we tested the prediction that a context rich in
multiple cues will facilitate sarcastic interpretation immediately and directly.

Rachel Giora

191

For instance, in Giora, Fein, Kaufman, Eisenberg, and Erez (2009) we showed
that context involving a frustrated expectation on the part of a protagonist did
not induce an expectation for a sarcastic utterance; nor did it facilitate sarcasm
interpretation compared to a context featuring a realized expectation. Instead,
the sarcastic utterances in both types of contexts took longer to process
compared to a context featuring no expectation, in which the appropriate
interpretation was salience-based.
But even when contexts were, in effect, shown to induce an expectation for
a sarcastic utterance, sarcastic interpretation was not facilitated immediately.
For instance, in Giora et al. (2007, Experiment 1), dialogic contexts were
shown to induce an expectation for a sarcastic irony by involving a sarcastic
speaker who uttered a sarcastic utterance twice: once in dialogue mid position
and once in dialogue final position. Results replicated previous findings.
Although a contextual expectation for a sarcastic utterance was induced,
processing the anticipated sarcastic utterances was slowed down compared to
their salience-based counterparts. Reinforcing such dialogues with explicit
marking (mockingly) did not affect the patterns of results (Giora, Yeari & Fein,
2012).
Similarly, when contextual expectation was manipulated by repeatedly and
exclusively exposing participants to contexts ending in a sarcastic utterance,
results were not affected: only salience-based interpretations were facilitated,
regardless of contextual misfit and length of processing time allowed (750,
1000 ms). In Giora, Yeari, and Fein (2012), this experimental design was
strengthened by providing participants with the information that the
experimenters were after sarcasm interpretation. Regardless, patterns of
results remained constant (see also Giora 2011). Multiple cues, whether
implicit or explicit, did not improve understanding of non-salient sarcastic
interpretations. Instead, only salience-based (often literal) interpretations were
activated initially, as predicted by the Graded Salience Hypothesis. Context,
then, is ineffective in blocking access of salient meanings and hence saliencebased interpretations early on.
2. In some of your papers, not only salience and context but also
(indirect) negation plays a fundamental role in explaining irony.
Recently you have also focused on explicit negation with regard to
sarcastic irony. Would you elaborate on the differences between the
two types of negation?

192

Humana.Mente Issue 23 December 2012

Whereas familiar verbal ironies have a coded sarcastic meaning, studying

unfamiliar verbal irony in terms of indirect negation deals with affirmative
utterances whose sarcastic interpretation is non-coded, but context
dependent, and thus needs to be constructed. (On different processing of
familiar and unfamiliar sarcastic ironies, see Filik, Leuthold, Wallington, &
Page, 2012; Giora & Fein, 1999a). As a result, even in the presence of highly
supportive contexts, processing unfamiliar ironies is taxing, compared to their
salience-based (often) literal interpretation (as discussed above). Indeed, there
is plenty of evidence demonstrating that unfamiliar utterances intended
sarcastically are slower to interpret appropriately compared to deriving their
salience-based albeit inappropriate interpretation. This evidence has been
accumulated by means of a variety of methodologies, both behavioral
(Akimoto, Miyazawa, & Muramoto, 2012 with regard to intentional irony;
Colston & Gibbs, 2002; Filik et al. 2012; Filik & Moxey, 2010; Gibbs, 1986;
Giora, Fein, Laadan et al., 2007; Ivanko & Pexman, 2003, Exp. 3; Pexman,
Ferretti & Katz, 2000), including brain damage (Giora, Zaidel, Soroker,
Batori, & Kasher, 2000), as well as automatic (Eviatar & Just, 2006; Filik et
al., 2012; for a summary see Giora, 1995, 2003).
In contrast, explicit negation induces sarcasm by default, independently of
contextual information. It allows comprehenders to activate sarcastic
interpretations of unfamiliar utterances directly, without having to go through
their salience-based (literal) interpretations first, which slows down derivation
of affirmative sarcasm. As shown by our recent studies, some novel negative
utterances of the form X s/he is not (supportive she is not), X is not her/his
forte (punctuality is not her forte), or X is not her/his strong point
(Thoroughness is not her strong point), involving no internal incongruity,
were interpreted sarcastically and rated as more sarcastic than their affirmative
counterparts, when presented in isolation. When embedded in strongly
supportive contexts, their non-salient sarcastically biased interpretation was
faster to activate than their salience-based literally biased interpretation (Giora,
Drucker, Fein, & Mendelson 2012; Giora, Livnat, Fein, Barnea, Zeiman, &
Berger in press; see also Giora, Fein, Ganzi, Alkeslassy Levi, & Sabah, 2005;
on negation as inducing default metaphorical interpretations, see Giora, Fein,
Metuki, & Stern, 2010).
These results are attributed to the role of negation as a low-salience marker,
highlighting meanings and interpretations low on salience by rejecting them

Rachel Giora

193

(Giora, Fein, Metuki et al., 2010; Givoni, Giora, & Bergerbest, in press). No
contemporary processing model, not least the Graded Salience Hypothesis,
can account for the priority of non-salient interpretations over salience-based
alternatives.
3. The Graded Salience Hypothesis you proposed explicitly avoids
abstract distinctions such as the literal/non-literal divide, and replaces
them with more fruitful concepts, such as salience, which is more finegrained and experimentally verifiable. You have further discarded the
distinction between literal and non-literal language with respect to
aesthetic effects. Can degree of salience also account for
pleasurability?
The literal and non-literal distinction is not entirely insignificant. However, it
cannot explain a number of findings which fail to distinguish literal from nonliteral language. For instance, it cannot account for the ease of processing of
familiar metaphors which is comparable to that of their salient or saliencebased, often literal interpretations (Giora & Fein, 1999b). Nor can it account
for the ease of processing of familiar ironies which is comparable to that of
their salience-based interpretations (Filik et al., 2012; Giora & Fein, 1999a).
In addition it cannot explain the slower reading times of salience-based literal
interpretations of highly familiar metaphors. Compared to their coded nonliteral meanings, which are high on salience, the literal interpretations of such
highly conventionalized metaphors are lower on salience and hence slower to
construct (Giora, Fein, Kronrod, Elnatan, Shuval, & Zur, 2004).
In addition, it can neither account for aesthetic effects induced by optimal
innovations which might be both literal and non-literal. According to Giora et
al. (2004), an optimal innovation is an expression which is novel (pinkwashing;
curl up and dye) but which also gives rise to a familiar meaning of a familiar
expression (whitewashing; curl up and die), so that the similarities and
dissimilarities between them may be considered. Although optimal innovations
take longer to process compared to the familiar expressions they activate, they
are rated as more aesthetic. In fact, they are rated more pleasing not just
compared to these highly familiar expressions which they deautomatize, but
also more pleasing than highly novel, or slightly altered counterparts,
regardless of degree of (non)literality. What can account for these results,
then, is not the literal non-literal distinction but degree of salience (see also

194

Humana.Mente Issue 23 December 2012

Giora, Fein, Kotler, & Shuval, in press; Shuval & Giora, 2009). The
literal/non-literal distinction (or even continuum, see Coulson & Van Petten,
2002) is not general enough to account for these findings.

REFERENCES

Akimoto, Y., Miyazawa, S. & Muramoto, T. (2012). Comprehension processes of

verbal irony: The effects of salience, egocentric context, and allocentric theory
of mind. Metaphor and Symbol, 27(3), 217242.
Campbell, J. D. & Katz, A. N. (2012). Are there necessary conditions for inducing a
sense of sarcastic irony? Discourse Processes, 49, 459480.
Colston, H. L. (2000). On the necessary conditions for verbal irony comprehension.
Pragmatics and Cognition, 8, 277324.
Colston, H. L. & Gibbs, R. W. Jr. (2002). Are irony and metaphor understood
differently? Metaphor and Symbol, 17, 5780.
Coulson, S., & Van Petten, C. (2002). Conceptual integration and metaphor
comprehension: An ERP Study. Memory & Cognition, 30, 958968.
Eviatar, Z., & Just, M.A. (2006). Brain correlates of discourse processing: An fMRI
investigation of
irony and conventional metaphor comprehension.
Neuropsychologia, 44, 23482359.
Filik, R., Leuthold, H., Wallington, K., & Page, J. (2012). Testing theories of irony
processing using eye-tracking and ERPs. Paper submitted for publication.
Gibbs, R. W. Jr. (1986). On the psycholinguistics of sarcasm. Journal of Experimental
Psychology: General, 115, 315.
Gibbs, R. W. Jr. (1994). The poetics of mind. Cambridge: Cambridge University
Press.
Gibbs, R. W. Jr. (2002). A new look at literal meaning in understanding what is said
and implicated. Journal of Pragmatics, 34, 457486.
Giora, R. (1995). On irony and negation. Discourse Processes, 19, 239264.
Giora, R. (1997). Understanding figurative and literal language: The graded salience
hypothesis. Cognitive Linguistics, 8(3), 183206.

Rachel Giora

195

Giora, R. (1999). On the priority of salient meanings: Studies of literal and figurative
language. Journal of Pragmatics, 31, 919929.
Giora, R. (2003). On our Mind: Salience, Context, and Figurative language. New
York: Oxford University Press.
Giora, R. (2011). Will anticipating irony facilitate it immediately? In: M. Dynel, (ed.).
The pragmatics of humour across discourse domains. Amsterdam: John
Benjamins, 1931.
Giora, R., Drucker, A., & Fein, O. (2011). Default ironic interpretations. Submitted
for publication.
Giora, R., & Fein, O. (1999a). Irony: Context and salience. Metaphor and Symbol,
14, 241257.
Giora, R. & Fein, O. (1999b). On understanding familiar and less-familiar figurative
language. Journal of Pragmatics, 31, 16011618.
Giora, R., Fein, O., Ganzi, J., Alkeslassy Levi, N., & Sabah, H. (2005) Negation as
mitigation: the case of negative irony. Discourse Processes, 39, 81100.
Giora, R., Fein, O. Kaufman, R. Eisenberg, D., & Erez, S. (2009). Does an ironic
situation favor an ironic interpretation? In G. Brne & J. Vandaele (Eds.),
Cognitive poetics. Goals, gains and gaps. (Applications of Cognitive
Linguistics series). Berlin/New York: Mouton de Gruyter, 383399.
Giora, R., Fein, O., Kotler, N., & Shuval, N. (In press). Know Hope: Metaphor,
optimal innovation, and pleasure. In: G. Brne, K. Feyaerts & T. Veale (Eds.),
Cognitive Linguistics Meet Humor Research . Current Trends and New
Developments. Berlin/New York: Mouton de Gruyter.
Giora, R., Fein, O., Kronrod, A., Elnatan, I., Shuval, N. & Zur, A. (2004). Weapons of
mass distraction: Optimal Innovation and Pleasure Ratings. Metaphor and
Symbol, 19, 115141.
Giora, R., Fein, O., Laadan, D., Wolfson, J., Zeituny, M., Kidron, R., Kaufman, R., &
Shaham, R. (2007). Expecting irony: Context vs. salience-based effects.
Metaphor and Symbol, 22(2), 119146.
Giora, R., Fein, O., Metuki, N., Stern, P. (2010). Negation as a metaphor-inducing
operator. In: L. Horn (Ed.), The expression of negation. Mouton de Gruyter

196

Humana.Mente Issue 23 December 2012

series "The expression of cognitive categories" under the general editorship of
Wolfgang Klein and Stephen Levinson, 225256.

Giora, R., Livnat, E., & Fein, O., Barnea, A., Zeiman, R., & Berger, I. (In press).
Negation generates nonliteral interpretations by default. Metaphor and
Symbol.
Giora, R., Yeari, M., & Fein, O. (2012). Sarcastic irony: Will expecting it make a
difference? Paper submitted for publication.
Giora, R., Zaidel, E., Soroker, N., Batori, G., & Kasher, A. (2000). Differential effects
of right and left hemispheric damage on understanding sarcasm and metaphor.
Metaphor and Symbol, 15, 63-83.
Givoni, S., Giora, R., & Bergerbest, D. (in press). How speakers alert addressees to
multiple meanings. Journal of Pragmatics.
Grice, H. P. (1975). Logic and Conversation. In P. Cole and J. Morgan (Eds.), Speech
Acts. Syntax and Semantics (Vol. 3). New York: Academic Press, 4158.
Katz, A. N. & Pexman, P. (1997). Interpreting figurative statements: Speaker
occupation can change metaphor into irony. Metaphor and Symbolic Activity,
12, 1941.
Pexman, P. M., Ferretti, T., & Katz, A. (2000). Discourse factors that influence irony
detection during on-line reading. Discourse Processes, 29, 201222.
Searle, J. (1979). Expression and meaning. Cambridge: Cambridge University Press.
Shuval, N. & Giora, R. (2009). Figurativeness, optimal innovation and pleasure.
Oryanut ve'Safa, 2, 111-127. (In Hebrew).
Utsumi, A. (2000). Verbal irony as implicit display of ironic environment:
Distinguishing ironic utterances from nonirony. Journal of Pragmatics, 32,
17771807.