Last draft forthcoming in:

Annalisa Coliva (editor), Eva Picardi on Language, Analysis and History, Palgrave, 2018.

Carlo Penco
Refusing to endorse: a must explanation for pejoratives

Abstract: In her analysis of pejoratives, Eva Picardi rejects a too sharp separation between
descriptive and expressive content. Carlo Penco reconstructs some of her arguments,
endorsing Eva’s criticism of Williamson’s analysis of Dummett and developing a suggestion
by Manuel Garcia Carpintero on a speech act analysis of pejoratives. Eva’s main concern is
accounting for our instinctive refusal to endorse an assertion containing pejoratives because it
suggests a picture of reality we do not share. Her stance might be further developed claiming
that uses of pejoratives not only suggest, but also promote a wrong picture of reality. Our
refusal to endorse implies rejecting not only a wrong picture of reality but also a call for
participation to what that picture promotes.

Key words

Pejoratives, sense and tone, assertibility conditions, Dummett, Frege,

Introduction

Since David Kaplan’s “The Meaning of ‘Ouch’ and ‘Oops’”, there has been a wide
amount of discussions on every side of pejorative expressions or slurs, with different
kinds of interpretations and new topics, like the problem of appropriation and
perspectival shift1. Picardi 2006, 2007 presents a set of suggestions concerning the use
of pejoratives and their relation to the content of what is said. Her stance is antagonist
towards a too easy “pragmatic” view of the matter, according to which a pejorative
doesn’t touch or is totally independent of what is said and only pertains to the level of
implicatures or presuppositions. On the contrary, Eva claims that the use of a
pejorative cannot be reduced to something always independent of the assertive
content, and that the use of pejoratives may pertain to the truth of the matter, given
that it predicates something false of the class to which it refers. Therefore, she would
be classified as belonging to the “semantic stance” proposed for instance by Hom
2008, 2010, 2012. According to the semantic stance, the derogatory content is part of
the meaning of the pejorative (e.g. “nigger” means something like “black and
despicable because of it”), and therefore a sentence containing a slur attributes an
empty property to the individual in question (Picardi 2006: 72), making the sentence
either false or deprived of truth value. Although she claims that the derogative aspect
of pejoratives is “part of a word’s literal meaning”, I think the morale of her papers

1
points towards a wider view on the role of pejoratives than the semantic one. I will
follow Eva's analysis of multi-proposition view (§1), her attempt to make derogative
terms impinge on truth conditions (§ 2), her reaction of Tim Williamson's criticism of
Michael Dummett (§3), her dubious attitude towards a presuppositional analysis (§4)
and eventually, in § 5, I conclude with a solution that seems to prompt from her
discontent with most answers to the problem of derogatory terms.

1. Sense, Tone and accompanying thoughts: a multiple propositions analysis.

In order to distinguish what a pejorative expression add to what is said, Kaplan

1999 distinguishes descriptives and expressives: the first describe what is or is not the
case; the second display what is or is not the case under a certain perspective or
attitude (two expressions may have the same information content and different
expressive content). According to Kaplan himself, this distinction is not so distant
from the Fregean analysis on the different contribution to content made by sense and
tone. Picardi 2006, 2007 discusses the Fregean distinction in relation to the use of
derogatory words. Frege considered tone or colouring as of pragmatic significance
and not pertaining to the truth-conditional content of an assertion (the assertoric
content). Frege's distinction antedates the distinction made by Paul Grice between
what is said (truth-conditional meaning) and what is meant (conventional or
conversational implicatures), as Picardi 2001 was one of the first to remark. A
standard example is given in Frege 1897 analysing the difference between:

(1) “That dog howled the whole night”

(2) “That cur howled the whole night”

According to Frege, the two sentences express the same truth-conditional content:
if the first is true, then the second also is true. Frege claims that (2), although
expressing the disapproval of the speaker, cannot be false if (1) is true. In fact, if we
thought that the aversion of the speaker was part of the content, the sentence should
be analysed as a conjunction of (1) “that dog howled the whole night” and something
like (3) “all dogs are despicable and ugly”. Assuming that (3) is false, the conjunction
of (1) and (3) would be false. Therefore, we could not accept the truth of sentence (2)
given that it is an expression of a complex proposition whose truth-value is false. But
we cannot assume that (2) is false while (1) is true, given that they refer to the same
state of affair. Supported by this argument, Frege distinguished between thoughts
whose assertion is expressed and thoughts that are not expressed, but only hinted at or
“suggested”, in order to influence the audience. Suggesting something using a
particular piece of the lexicon to refer to an individual does not concern the problem
of truth and belongs to the realm of colouring or tone, which pertains to pragmatic
aspects of language (Frege 1897, 1879 § 3). The solution to the analysis of pejoratives

2
seems straightforward: conveying something suggested and not explicitly asserted, a
sentence with a pejorative does not concern what is said, but what is meant, or the
conventional implicature. Picardi is not happy with this solution and tries different
ways to go beyond it.
First of all, Picardi criticises Frege for assuming too easily that “dog” and “cur” are
coreferential. If so, the two terms should require substitutability salva veritate, but
there are counterexamples:

“To his neighbor's utterance ‘That cur howled all night’, the owner of the dog may retorts,
‘That dog is not a cur’, but plainly he is not asserting that his dog is not a dog. Possibly, all
curs are dogs, but not all dogs are curs. All that Frege is entitled to say is that there are
contexts of utterance in which the difference in meaning between “cur” and “dog” makes no
difference to truth-conditions of what is said, whereas there are other contexts in which the
difference is salient.” (Picardi 2006: 62)

The main claim given by this example is that we cannot take it for granted that a
neutral term and a pejorative have always the same extension. However the
disagreement between two speakers here does not grant the conclusion; in fact the two
expressions (the neutral one and the derogative one) have the same extension in the
mind of the dog hater, and when they are used to refer, the reference is normally
successful because the interlocutor easily gets what the speaker has in mind. Saying
that it is false that all dogs are curs is an expression of disagreement on the different
uses of the words, not on the factual truth of the assertion of the speaker (whether the
animal – in whichever way you want to refer to it – howled all night). It seems to me
therefore that this argument is not strong enough to avoid the conclusion that
assertions containing pejoratives have the same truth-conditional content than
assertions with neutral terms, insofar the pejoratives are used to refer and are
understood as such.
What about a multi propositional analysis? Relying on a long tradition of research
(Kent Bach, Robyn Carston, Francois Recanati), Picardi claims that the difference
between “what is said” and “what is conventionally implicated” is not sharp enough
to decide without doubts when something belongs to the content of an assertion and
when it does not, given that the choice may depend on the point of the assertion. A
possible solution might be to translate the sentence (2) with an explicature (or a free
enrichment), as:

(4) “That dog, which is despicable and ugly because of it, howled all night”.

With this peculiar rendering, we might answer to the Fregean strategy for which it
is counterintuitive to take (2) as false considered as a conjunction (“that dog howled
and all dogs are despicable and ugly”). In fact, with (4) interpreted as an explicature
of (2) we would really have a different proposition from (1) and we may admit – in
this case – that the truth of the content of the main assertion (that the dog howled all
night) is affected by the truth-value of the relative clause. This might be a possible

3
“multi-propositional” solution of the relevance of pejoratives to what is said.
We may claim therefore that the use of the pejorative is intended to imply that the fact
that dogs are despicable is a reason or a cause why they howl all night or vice-versa.
This last point seems the best way to explain Picardi’s criticism of Kaplan’s analysis
of Frege’s accompanying thoughts (Nebengedanke) with which she shares much, but
not all. The discussion starts with Frege’s example:

(5) “Napoleon, who recognized the danger to his right flank, personally led his guards
against the enemy position”

Picardi remarks that Frege realizes that “the clause expresses more through its
connection with another than it does in isolation” (Frege 1892: 47) A relative clause
cannot be represented always with the same syntactic form. We have different ways to
compose an accompanying thought expressed by a relative clause, and while some of
them make it independent of the whole, other really affect what is said with the
composed sentence. In the case under discussion, Picardi suggests that what is said
may be conceived as inserting a third thought, that is

(6) “the recognition of the danger is a reason why Napoleon led his attack”.

In this case, it becomes apparent that the accompanying thought may be part of
what is said, given that it impinges on the truth-condition of the composed sentence.
Picardi really makes a case about that. The conclusion is that a pejorative may be
treated as prompting a further proposition that cannot be conceived just as a
conventional implicature, but as an explicature – that is part of what is said –
presenting a point, such as individuating the reason explaining the content of the main
clause: seeing the danger of his side is a reason for Napoleon to attack, or being ugly
and despicable is a cause for the dog to howl all night – maybe because despicable
animals just do that.
Although this is a possible analysis of pejoratives that makes justice of the idea
that an assertion containing a pejorative may be just false, Picardi eventually rejects it.
In fact she claims that the idea of a specific completion of a sentence because of a
pejorative is not sound; following Sainsbury 2001, she claims that what is relevant
with a sentence with expressive content is its lacking of specificity, and therefore the
sentence “should not be construed as directed to an elliptical proposition that awaits to
be spelled out in full” (Picardi 2006: 54).

2. Do pejorative really impinge on truth conditions?

Without the help of explicatures or free enrichment, however, it becomes difficult

to claim that pejoratives pertain to the assertoric content, to what is said. Yet Eva,
criticising Kaplan’s too sharp separation between expressives and descriptives, attacks
the rendering of this distinction made by Potts 2005, 2008, who considers expressives
as conventional implicatures. According to Potts the expressive meaning of a lexical

4
item is independent of its descriptive meaning and therefore plays no role in
determining the truth conditions. The main point of disagreement with the above
distinction concerns the claim of independence of the expressive content. Eva’s
criticism of the idea of independence of expressives works on a basic question:
How can we consider the correctness of a reported speech in case the original
speech contained a pejorative?
At first sight, reporting an utterance with a pejorative like “that cur howled all
night”, a lover of dogs would probably make a report of the kind: “x said that that dog
howled all night”, abstaining to use the pejorative term, but still thinking to have
made a correct report of what happened, preserving at least the truth of the matter. But
not everybody would agree of the correcness of the report. Eva refers to Bach 2006
whose argument for claiming that conventional implicatures belong to what is said is
that they fail the indirect speech test; if you report John’s having said “Mary is pretty
but intelligent” as “John said that Mary is pretty and intelligent”, Bach doubts that
you have made a correct report. In the reported speech we should make it clear that
the speaker intended a contraposition between the first and the second property. We
should have an enriched proposition that could make explicit the content of the
contrast. We have seen however that this is not the path followed by Eva. Which
means are still available to fight a analysis of pejoratives based on the idea of
conventional implicature?
Instead of following the multi-propositional analysis, Eva pinpoints another
possible problem: the relevance of what is the “at-issue” content or the question under
discussion (QUD). On this point she makes an example purporting to show the
difficulty of sharply separating the assertoric content from the implicated content:

“Whether my leaving out this piece of information renders my report wrong or simply inaccurate
depends on what was the main point of the utterance on the given occasion. And this, in its turn,
depends on the audience I am addressing and on the focus of the conversation: in the course of an
investigation that aims at discovering the culprit of evil deeds against dogs in a certain neighbourhood
it may be useful to give a literal report of what the people involved say concerning dogs. In a different
context, the report may be less accurate, if, for instance, our interlocutor simply wants to find out what
a notoriously nagging neighbour was complaining about.”

This is an ingenious effort to defend the claim that pejoratives enter the question of
truth. But it seems to me that here we have two questions: one question concerns the
facts described by the report, another question concerns the facts concerning the
psychology of the speaker: if we are looking for a devious assassin of dogs, reporting
the specific lexical item impinges on the latter. We might have evidence, although
inconclusive, of the speaker's tendency to perform crimes against dogs. Here,
therefore, a literal report may be of fundamental importance to denounce the speaker.
However, again, derogatory conceptualizations do not change the “strict” truth-
conditional content of a description of a state of affairs. In fact, in this case, the truth
evaluation concerns (the fact of) which words the speaker used, not which facts have
been reported about the behavior of dogs.
In fact, “that cur” is a complex demonstrative, whose main role in the sentence is

5
the identification of the referent; we may think that it presents the referent in a wrong
way and, from this perspective, is not too dissimilar to a misdescription. A speaker
may make a referntial use of an inaccurate definite description assuming the hearer
may understand the intended referent although the description is false of it, or at least
defective (see also Penco 2010, 2017). From the point of view of truth conditions,
both misdescriptions and pejoratives may be considered defective but still able to
make the hearer correctly understand the referential intentions.
Picardi (2006: 67) is well aware of the problem, and she refers to Donnellan on this
point. Her use of the similarity with Donnellan’s cases helps to point out the
difference between the case of misdescriptions and the case of pejoratives. In case of
misdescriptions like “the man drinking champagne” (while he is drinking mineral
water), there is no harm in using a defective or inaccurate or wrong definite
description if your referential intentions are understood. On the contrary this does not
happen with pejoratives. While with misdescriptions we are in front of a factual
mistake, whose correction is easy to accept (“the person you are referring is not
drinking chanpagne but mineral water”), in front of a derogatory term you cannot
simply change the term and be happy, because you are facing a strong disagreement:
what is wrong from the point of view of dog lovers, may be strongly believed by the
dog hater, who would not recede from his conceptualization of that class (dogs are
despicable because of being dogs and blacks are inferior because of being black).
Besides, given certain circumstances, I may easily report what a speaker said using
the same misdescription to make myself understood, but this is not so with pejoratives.
In fact I might feel uncomfortable using a pejorative term, on whose grounds and
consequences I don’t agree. But this does not mean that I don’t understand what the
speaker said and I have to distinguish between the facts of the matter described and
the point of view and attitude of the speaker.
As Dummett (2007: 527) says, commenting of Picardi’s paper, “the use of a
pejorative expression certainly cannot be said to affect the truth-value of an utterance;
it affects its property. But, for the same reason, it also cannot simply be explained as
affecting the tone of the utterance, or as attaching an implicature to it.” The offensive
character of certain terms, Dummett claims, should be accounted for by “the license
they give their user to draw inappropriate consequences”. In conclusion, we cannot
use the test of reported speech to claim that pejorative impinge on the truth of the
matter, although we may still take our awareness on “the tacit commitments we would
undertake in accepting a certain way of referring to people or actions” as a ground to
refuse to endorse an assertion (Picardi 2007: 507).
The question seems to shift from the truth-value of an assertion concerning a state
of affair (what did the dog do during the night?) to the justifications and consequences
of assertions containing a pejorative. If truth conditions are not affected, pejoratives
certainly affect assertibility conditions. Different lexical items are connected with
different justification and consequences, and using them obliges us to explain why we
have used them and commits us to the consequences of what they mean. We are
entering another kind of problem, that touches upon the meaning of pejoratives: while
it seems that truth conditions are affected only by the referents of pejoratives,

6
assertibility conditions may be affected by their meanings. On the meaning of
pejoratives, Picardi is very near to Dummett’s classical analysis and contrasts
Williamson’s criticism of this analysis.

3. Pejoratives as dealing with truth/assertibility conditions

Picardi 2006, 2007 looks back at Dummett's discussion on the logic of pejoratives.
Dummett was interested in the logical role of pejoratives and in the logical
motivations to reject their use. His claim was clear and simple: a pejorative like
“Boche” – used to refer to Germans implying that Germans are more prone to cruelty
than other Europeans – would produce a non conservative extension of the language
where the word was not present: the use of “Boche” would permit inferences and
conclusions that would not be permitted in the language missing the pejorative
(Dummett 1981: 454).
Dummett gives Introduction rules and Elimination rules for the term and shows
how they permit conclusions impossible to be derived without the term. The
Introduction Rule for “Boche” (or its condition of application) would simply be
something like

x is German
–––––––––––––
x is Boche

But the consequences of application embed the following inference (that might be
considered the Elimination rule for “Boche”):

x is Boche
––––––––––––––––––––––––––––––––––––––
x is more prone to cruelty than other Europeans

Now, if we accept the Introduction and Elimination rules for Boche we should
derive something of the following: Angela Merkel is German therefore is Boche, and
if Angela Merkel is Boche, therefore, she is more prone to cruelty than other
Europeans. We could not to derive this conclusion from our lexicon only if, following
the elegant attitude of Oscar Wilde, we did not include the lexical item “Boche” in our
dictionary.
Williamson launches an attack on radical inferentialism and defends a radical
referentialist framework (although he recognizes that there are intermediary positions
that might escape his criticism). He criticizes Boghossian (2203: 241-2) according to
whom “plausibly, a thinker possesses the concept Boche just in case he is willing to
infer according to [Dummett’s rules]”, with the following short argument:

“Since understanding the word ‘Boche’ (with its present meaning) is presumably sufficient
(although not necessary) for having the concept that ‘Boche’ expresses, it follows that a

7
willingness or disposition to reason according to Dummett’s rules is equally unnecessary for
having that concept.” (Williamson (2009: 8-9)

This claim is correct, and maybe Boghossian went a bit too far. Let us assume that
understanding a concept is understanding its introduction and elimination rules.
Mastering those rules is not to be identified with willingness to follow them, but with an
implicit knowledge of them. Understanding the meaning is understanding what it is or
what it would be using that inference, even without explicitly doing so or even
rejecting to endorse it. Williamson may accept that, but then – he would ask – which
is the difference between an inferentialist and a referentialist account of understanding
as a “practical” ability if we cut off the actual disposition to reason according to the
rules? The answer is that, although there is no difference in “practical” ability,
inferentialists are not content of getting the referent right: they require making the
inferential connections explicit. Explicitly rejecting to use a term is exactly the point
of the difference between a referentialist and an inferentialist view. You refuse to use
a term because you reject the possible consequences of its use; referential rules on the
other hand just point out that the reference of “German” is the same of the reference
of “Boche”; the two classes are the same class:

“differences between ‘Boche’ and ‘German’ apparently play no role in determining reference, and
so make no difference to the way in which the terms contribute to the truth-conditions of sentences in
which they occur, a Fregean might even count ‘Boche’ and ‘German’ as having the same sense. Frege
himself gives just such an account of another pejorative term (1979: 140): ‘cur’ has the same sense and
reference as ‘dog’ but a different tone.” (122)

Here we are. Again on Frege, and our examples of pejoratives for “dogs”! As we
have seen, the main role of pejoratives does not concern just the role of reference
fixing, as in case of definite descriptions, but their role in suggesting inferences to be
accepted (conventional implicatures). On this point, Williamson himself concedes the
idea that conventional implicatures (something derivable and therefore linked to an
inferential structure) of expressions like “Boche” are “part of their meaning in a broad
sense of meaning”. But, if we accept an idea of (broad) meaning as dealing with the
inferences connected with an expression, then Dummett’s proposal is not
incompatible with a view of pejorative as triggering a conventional implicature or a
presupposition.
A way to interpret Dummett’s treatment of pejoratives is then to consider it as a
clarification of the rules behind what is expressed and not stated, rules that should be
followed if one accepts the conventional implicature connected with the use of the
pejorative. Accepting a pejorative, we accept a network of inferences, a set of beliefs
that the pejorative brings with it. Using an assertion with a pejorative is not only
saying something true with a bad psychological surrounding: it is accepting the
consequences connected to the inferential meaning of the expression. We are back to
the conclusion of the previous discussion: we understand the intended referents of
singular terms or complex demonstratives like “that dog” or “that cur”, and we
understand to which classes predicates like “German” or “Boche” refer; although

8
sentences containing them may have, by substitution of coreferentials, the same truth
conditions, they certainly haven’t the same assertibility conditions; in fact, to have the
same assertibility conditions they should also have the same ground for justification,
and we may claim, from our perspective, that nobody is justified to use “cur” or
“Boche” given that those terms imply consequences that we disagree about.

4. Problems of pejoratives as presupposition triggers

If a conventional implicature can be considered part of the “broad” meaning of an

expression, then it seems that Picardi (2007: 508) herself makes a too strong contrast
between “the decision to construe the explicit derogatory ingredient as a conventional
implicature” on one hand and the idea of construing the derogatory ingredient “as
constitutive of word meaning” on the other. The two aspects are not antagonist:
considering the derogatory ingredient as conventional implicature implies that its
broad meaning is connected with the inferences that are derivable by its use and are
suggested as “calculable” implicature.
However, speaking of inferences syntactically plugged into the lexicon,
conventional implicatures may not be the best solution for treating pejoratives. The
other solution is treating them as triggering presuppositions. Actually conventional
implicatures pass the S-Family test of presuppositions: they stand also when an
assertion is made in negative, interrogative and modal form (“that cur didn’t howl all
night”, “did that cur howl all night?”, “that cur might have howled all night”).
Let us then see what happens when treating pejoratives as presupposition triggers
(for a defence of a presuppositipnal account see Schlenker 2007, Macia 2014 and
Cepollaro 2015). Under this perspective, the use of a pejorative presupposes the set of
beliefs that the pejorative intends to implicate. This should be coherent with the
classical view of presuppositions for which an utterance of a sentence is appropriate
if the presupposition is shared by the participants of a conversation (Stalnaker 1974).
We may consider the use of pejorative not appropriate if we do not share the
presupposition; therefore, as Picardi (2007: 507) claims “we may abstain from
accepting a statement made by others because we are aware of the tacit commitments
we would undertake in accepting a certain way of referring to people or actions”.
This claim is perfectly adequate to a presuppositional analysis, and it seems to me
that presuppositional analysis and the difference between truth conditions and
assertibility conditions come hand in hand; abstaining from endorsing a statement
means rejecting the justifications or the intended background for its assertion.
A presentation of a presuppositional analysis might also be framed in Kaplan’s
terminology. Kaplan 1999 was interested in the informational content that can be
derived by expressives; in doing so he attempted to clarify the rules of correctness of
expressives and the correspondence of informational content given by expressives and
by descriptions: the same semantic information can be given with an expressive mode
(ouch, oops, hurray) or with a descriptive mode (“I am in pain” or “I have just
observed a minor mishap” or “I am in state of joyful elation”). Kaplan describes the
problem of giving the rules for correct application of expressives. And we might say

9
that the felicity condition of the use of a pejorative is that (1) the person actually
believes the information content expressed by the pejorative and (2) has the correct
attitude or emotion towards the class described by the pejorative. Utterances of “that
Boche run away” or “that cur howled all night” are appropriate only if the speaker
really believes that Germans are cruel as such or that dogs are despicable and ugly as
such and has the appropriate emotion of distaste or dislike (see also Carpintero 2017)
This is what presuppositional analysis amounts to.
However, the presuppositional analysis of pejoratives leaves unanswered some
questions like the following:
(1) In using a slur in a re-appropriation case, people do not share the prejudice (the
belief) attached to the term, yet it seems that their use is appropriate.
(2) When people who have racist prejudices use the derogatory term we should say
that their use is perfectly "appropriate" because they share the beliefs supporting that
use, yet it sounds awkward, although correct for the theory, to say that the uses of
derogatory terms are somehow "appropriate”.
(3) when a presupposition is expressed it ceases to be a presupposition and it is
normally accepted while the presupposed content of a slur is typically a matter of
disagreement when explicitly stated.
Leaving (1) and (2) to the reader, let us see the problem with (3). Accommodation
(the process through which people accept presuppositions that do not belong to the
common ground) is not as normal as it is in standard cases (where, as Lewis says,
presuppositions “spring into existence making what you say acceptable after all”). A
non-xenophobe, or a non-racist, or a friend of dogs, would not easily accommodate
the presupposition in a sentence that uses a pejorative. He would probably say, “Hey,
wait a moment! Do you think that Germans are more prone to cruelty than other
Europeans? It is not true” or “hey wait a moment! Do you think that all dogs are
despicable? That’s false”. The problem arises because the presupposition triggered by
a pejorative represents a content on which there may be very strong disagreement.
The main defect of presuppositional analysis is that it leaves something out;
offensive or derogatory terms does not only pertain to the content of their
presuppositions (and eventually the emotional attitude going with it); they also
involve actions and commitments undertaken in their use.

5. Pejoratives as higher order utterance modifiers

Eva makes a remark on Frege’s view of the derogatory ingredient attached to the
word “cur”:

“[according to Frege] in the given context the choice of “cur” instead of “dog” has the
value of an exclamation, and, one may add, could be rendered syntactically by means of an
exclamation mark, much as assertoric force is rendered by means of a vertical stroke. Frege
held that assertoric force only shows itself with the help of a suitable notation, but is not
located in any part of speech in particular. Its scope is the whole utterance, not a particular
segment of it. The function of an interjection mark encapsulated, as it were, into the meaning

10
of “cur” in the specific utterance is to disclose the attitude of the speaker towards the matter at
issue. It presents the dog as ugly or unpleasant from the speaker’s perspective; however, as
Frege remarks, the dog itself may very well be a handsome representative of its race. But this
circumstance does not render the use of the interjection incorrect, for in uttering as he did, the
speaker might have wished to disclose his attitude of dislike or fear of dogs in general, not of
this dog in particular.” (Picardi 2006: 62-63)

Eva here refers also to Kaplan, who distinguishes between “truth simpliciter” and
“truth with an attitude”, but – as we have seen – she does not agree to treat the
expressive content merely as something propositional: “tone need not be expressible
by means of a, let alone one specific, full-fledged proposition" (Picardi 2007: 503).
But which kind of non-propositional aspect can be conveyed by a derogatory
expression?
Eva attributes the main reason for accepting the Fregean suggestion of colouring as
higher order utterance modifier to the fact that it detaches the notion of colouring
from mere psychological significance.2 A pejorative may impinge on the level of
speech acts, on their felicity conditions or justification (or assertibility) conditions.
This is a central point to be clarified.
Eva oscillates between two alternatives often connected: a multi-propositional
account and a higher order account, both of which she tends to disregard. But I think
she has been too quick with disregarding the idea of higher-order account, maybe
because too strictly connected with the Gricean view. Speaking of higher order
account we typically tend to consider the contribution of some words (expressions
like “but” or pejoratives) as parasitic on a ground floor or central speech act (Grice
1989: 361-2)3. But the idea of higher order modifier is not exhausted by Grice’s view
of implicatures (nor by the presuppositional account). An alternative view may be
defined for treating pejoratives as higher order modifiers that are not just parasitic
aspects. Still keeping pejoratives as connected with a set of inferences (presupposed
or implicated), Carpintero 2015, 2017 tries to give them a further role in the context
of dialogue. The novel point that Carpintero makes it where to insert the role of
pejorative in the dialogue: not only as part of the content or as presupposed
propositions, but as constraints on the context of dialogue. The main consequence of
accepting derogatory expressions is the implicit acceptance of their presuppositional
content, given by tacit accommodation. Tacit accommodation implies tacit
undertaking of a network of inferences and commitment to the consequences.
The main point made by Carpintero is, therefore, that common ground cannot be
defined only in terms of shared propositions, but also in terms of different
commitments towards those propositions; that’s why we feel so unconfortable when
we are included in a conversation where people use pejoratives on whose
stereotypical inferences we strongly disagree. Already Stalnaker claimed that we have
different attitudes towards the contents of the common ground (see Domaneschi et
alia 2014). But Carpintero’s point is stronger and can be summarised by the claim that
our common ground is made not only of propositions and propositional attitudes but
it also concerns attitudes linked to illocutionary forces, which is a further level of

11
pragmatic commitment.
Saying that the use of pejorative is linked to illocutionary force is a fundamental
step, shared by many others. Langton 1993, 2012 calls “speech acts of subordination”
those speech acts used to classify a target class as inferior, legitimate discrimination
and deprive it of rights. But her examples are basically explicit acts of subordination
like “Blacks are not permitted to vote”, where the act is a kind of “verdictive" and the
speaker has authority to do that because he is in a social position that allows him to
perform the act. Besides, in case the speaker has no authority, the accommodation of
the presupposition (given by the failure to question the speaker) would confer
authority to the speaker herself, as suggested by Maitra 2012 (McGowan 2004, 2009
speaks of "conversational" exercitives that, differently from Austin's, do not require
uptake from the hearers).
However most of the examples of this literature concern explicit and direct acts of
subordination (like the above quoted “Blacks are not permitted to vote”), or hate
speech that is characterised, among other things, by being directly addressed to the
individuals whom they insult (see also Hornsby 2003: 297). On the contrary our
examples (following Frege’s example with “cur”) concern the use of derogative
words in descriptions of facts or in questions, where the pejorative is part of a
descriptive content of a phrase (complex demonstratives, definite descriptions) whose
main function is to pick a referent. How to describe the subtler way in which the
insertion of a pejorative in a normal description of facts changes – to use Lewis'
terminology – the conversational score?
A first solution is to think of an indirect speech act: by putting a question such as
“do you know what time is it?” I make a request; by describing a possible situation
such as “I will not miss your date” I am making a promise. Following the analogy, we
may say that by describing a situation with “that Boche run away”, or asking “have
you seen that Boche running away?” I hereby promote discrimination and legitimize
behaviour of discrimination. We may think that the preparatory condition is not
satisfied. If asked: “can you tell me the time?” I may answer: “Sorry, I have no
watch”, making it clear that the preparatory condition of the request is not satisfied.
Analogously, if asked: “have you seen a nigger running away?” I may answer: “Sorry,
for me there are no nigger”, because the preparatory condition to the act of
subordination requires that black people are inferior as such. But the analogy is not so
clear: an indirect speech act is typically a speech act of a kind that is given by a
speech act of a different kind. By a question we make a request, by a description we
make a promise; in case of assertions or questions containing a derogatory term we
are still making an assertion or a question. Saying: “hey, wait a moment; he run away,
but he is not a nigger” (as with rejecting a presupposition), we correctly answer the
main speech act; while we cannot say to a question like “do you know the time”
something like, “yes, we do, but unfortunately I have no watch”. The strategy of
indirect speech act after all seems not to be a viable analysis.
A second possible answer, that seems to be more coherent with the main trend in
contemporary discussion, is that speech acts with derogatory terms contain a peculiar
adjunctive force: with the same utterance, we make two kinds of speech acts at the

12
same time (Kissine 2013: 197): assertions, questions, commands, and other speech
acts can be understood as such, and at the same time, when containing a derogatory
term, they are at the same time acts of “subordination”. And also, we have two
contents: the (description of the) objective state of affairs (a person who runs away)
and the (promotion of a) derogatory viewpoint concerning the individual and the
group they belong to. Langton 2017 presented a similar idea at the ECAP Conference
in Münich, speaking of “Blocking as Counter-speech” (e.g. you may assert something
normal and at the same time, through a presuppositional trigger, convey something
else like in “Even John could win!”)4. We may conclude that speech acts that contain
derogatory terms (or other subtle means to give a diminishing perspective on the
target) promote and legitimize subordination or other negative attitudes towards the
referent class. And, most of the time5, the subordination is derived by the use of a
predicate that is false of the class in question, for instance, because “the complex
properties indicated by racist words are not instantiated” (Picardi 2006: 68).
Can we be content in saying that with pejoratives we make two speech acts at the
same time? The idea of a speech act of subordination is still a pointer towards an idea
to be refined, and we might distinguish levels of subordination, and also other kinds
of acts depending on different kinds of pejoratives or on different targets or different
social roles (on which se e.g. Popa-Wyatt 2017). The essential feature, besides the
actual contempt or disregard of the target, is normally taken to be that in using a
pejorative we act to promote and legitimize subordination. What is not yet clear is
how this promotion is realised.
My suggestion is that who intentionally6 uses derogatory terms looks for company,
for sharing the prejudice and avoiding feeling alone. His speech act constitutes a call
for joint responsibility, asking for sharing an attitude towards the derogatory content,
indirectly creating a context of commitment to certain behaviour against the target.
This is why rejecting to endorse an assertion or answering a question containing a
pejorative is the fundamental reaction and avoids the trick of the use of derogatory
terms; on the one hand it seems that the racist (or the dog hater) is just stating some
facts and therefore we are ready to accept or reject the truth of the matter; but in
stating some facts with a certain terminology the racist (or the dog hater) is
desperately asking for approval of his behaviour and his way of life, and for sharing
his positive endeavour to promote this behaviour and way of life.
Summarizing, the use of a derogatory term in a normal speech act gives the act a
new feature, besides promoting discrimination or subordination: it is a call for joint
responsibility that commits co-conversationalists to participate in the actual
subordination and deprecation of the individuals or classes defined with a pejorative.
Therefore, the use of pejoratives is not just a question of informational content, or of
tacit presuppositions, but it is promotion and legitimization of that content through
tacit joint acceptance. In accepting a presuppositional content we ourselves turn to be
promoters of that content, and not only making as if we believe it.
This seems to me a fairly acceptable rendering of the central core of Eva's analysis
concerning the relationship between assertion and endorsement:

13
 I may refuse to endorse an assertion because its wording suggests a picture of reality that I do
not share.” (Picardi 2006: 62)

The central point of the refusal to endorse is rejecting the call for joint
responsibility and leaving the racist alone. And probably, under this “illocutionary”
view, we are allowed to say something stronger: the use of words not only “suggests”
a picture of reality, but also actually “promotes” it.

Summary

The connection between the speech act and the set of inferences connected (either
because of implicatures or because of presuppositions) with the pejorative expression
builds up a new challenging problem on the relationship between truth conditions and
assertibility conditions, and this seems to be the most relevant suggestion left by
Eva’s paper. Rejecting to endorse an assertion containing derogatory terms aims both
at preventing the derivations of other assertions whose content would entail what we
regard as false, and at preventing the promotion of what we considered wrong
attitudes towards the object of contempt.7

References

Bach, K. 1999 “The Myth of Conventional Implicatures”, Linguistics and

Philosophy 229, 327-66.
Bianchi C. 2014 “The speech acts account of derogatory epithets: some critical
notes”, Liber Amicorum Pascal Engel, ed. J.Dutant, D.Fassio, A.Meylan, Université
de Genève.
Bianchi C. 2015 “A plea for an experimental approach on slurs”, Language
Sciences
Carpintero: see Garcia-Carpintero
Cepollaro B. 2015 “In defense of a presuppositional account of slurs”, Language
Sciences, 52:36-45.
Cozzo C. 2002, "Does Epistemological Holism lead to Meaning Holism?" in Topoi,
1-2: 25-45.
Dummett M. 1981, Frege, Philosophy of Language, 2nd edition, , London:
Duckworth.
Dummett M. 2007 “Reply to Eva Picardi”, in Schilpp & Hahn (eds), The
Philosophy of Michael Dummett (LLP), LaSalle: Open Court.
Frege G. (1879) Begriffsschrift, L. Nebert, Halle.
Frege, G. 1897 “Logik” in Nachgelassene Schriften, edited by H. Hermes (et alia),
Hamburg: Meiner. 1969; Engl. Translation by P. Long and R. White, Posthumous

14
Writings, Oxford, Blackwell, 1979.
Garcia Carpintero M. 2015 “Contexts as Shared Commitments”, Frontiers of
Psychology, 6
Garcia Carpintero M. 2017 “Pejoratives, Contexts and Presuppositions” in Lecture
Notes in Artificial Intelligence, 10257: 15-24
Grice, P.H. 1967 “Logic and Conversation”, chapter 2 of Grice 1989, Studies in the
Way of Words, Cambridge Mass: Harvard University Press.
Grice, P.H. 1989 "Retrospective Epilogue", in Grice (1989) Studies in the Way of
Words, Cambridge (Mass): Harvard University Press: 339-385
Hom, C. 2008 “The Semantics of Racial Epithets”, Journal of Philosophy 105,
416-40.
Hom, C. 2010 “Pejoratives”, Philosophy Compass 5(2), 164-185.
Hom, C. 2012 “The Puzzle of Pejoratives”, Philosophical Studies 159, 383-405.
Hornsby J. (2001) “Meaning and uselessness: how to think about derogatory
words”, in Midwest Studies in Philosophy, XXV (pp.128-141)
Kaplan, D. 1999 “The meaning of ouch and oops. Exploration in the theory of
meaning as use”, Unpublished.
Langton, R. 2012 “Beyond Belief: Pragmatics in Hate Speech and Pornography”,
In McGowan and Maitra (Eds.) What Speech Does, Oxford: Oxford University Press.
Langton, R. 2017 “Blocking as Counter-Speech” paper presente at ECAP9,
Conference of the European Society for Analytic Philosophy (München 2017). The
paper is forthcoming in D. Harris, D. Fogal, M.Moss (eds), New Work on Speech
Acts. New York: Oxford University Press.
Kissine, M. 2013 From Utterances to Speech Acts, Cambridge: Cambridge
UniversityPress.
Macià, Josep 2014 “Expressive Meaning and Presupposition”, handout for a talk at
the Names, Demonstratives, and Expressives conference in Gargnano, Italy.
Maitra, I. 2012. Subordinating Speech. In Speech and Harm: Controversies Over
Free Speech, eds. Ishani Maitra and Mary K. McGowan, 94-120. Oxford: Oxford
University Press.
McGowan, M.K. 2004, “Conversational Exercitives: Something Else We Do With
Our Words”, Linguistics and Philosophy 27 (1): 93–111.
McGowan, M.K. 2009. “Oppressive Speech”, Australasian Journal of Philosophy
87 (3): 389-407.
McGowan, M.K. 2012, “On ‘Whites Only’ Signs and Racist Hate Speech: Verbal
Acts of Racial Discrimination”, in Ishani Maitra and Mary K. McGowan (eds.)
Speech and Harm: Controversies Over Free Speech, Oxford: Oxford University
Press: 94-120.
Mulligan, K. 1998 “From Appropriate Emotions to Values,” Monist, 81(1): 161–88.
Penco C. 2017 “Donnellan's Misdescriptions and Loose Talk” in K.Korta & M.
Ponte, Reference and Representation in language and Thought, Oxford U.P.
Penco C. forth. Prejudice and Presupposition in Offensive Language, in Nordic
and Mediterranean (forthcoming)
Picardi, E. 2001 “Compositionality”, in G. Cosenza (ed.), Grice’s Heritage

15
Turnhout: Brepols, pp. 52-72.
Picardi E. 2006 “Colouring, multiple propositions and Assertoric content” in
Grazer Philosophische Studien, 72: 49-72.
Picardi, E. 2007 “On Sense, Tone, and Accompanying Thoughts”, in Schilpp &
Hahn (eds), The Philosophy of Michael Dummett (LLP), LaSalle: Open Court.
Popa-Wyatt M. and Wyatt J.L. 2017, Slurs, roles and power, Philosophical Studies,
2017 (DOI 10.1007/s11098-017-0986-2)
Sainsbury, M. 2001 “Two ways to smoke a cigarette”, Ratio, 14: 386-406.
Schlenker, P. 2007 “The Expressive Dimensions”, Theoretical Linguistics 33(2),
237-245
Stalnaker, R. 1978 “Assertion,” in P Cole (ed.) Syntax and Semantics 9, New
York: Academic Press, 315-332.
Stalnaker, R. 1978 “Assertion,” in P Cole (ed.) Syntax and Semantics 9, New
York: Academic Press, 315-332. Also in R Stalnaker, Context and Content, Oxford
UP, Oxford, 1999, pp. 78-95, to which I refer.
Stalnaker, R. 2002 “Common Ground”, Linguistics and Philosophy 25: 701–721.
Stalnaker R. 2014 Context, Oxford: Oxford University Press.
Williamson, T. 2003 “Understanding and Inference”, Proceedings of the
Aristotelian Society, pp. 249-293.
Williamson, T. 2009 “Reference, Inference and the Semantics of Pejoratives", in J.
Almog & P. Leonardi, The Philosophy of David Kaplan, Oxford: Oxford U.P. (137-
158).
Wittgenstein, L. 1953 Philosophische Untersuchungen/ Philosophical
Investigations, edited by G.E.M. Anscombe and R. Rhees, trans. G.E.M. Anscombe,
Oxford: Blackwell.

1
A short summary of different perspective is given in Bianchi 2014 and Bianchi 2015
(also with reference to experimental approaches) and Cepollaro 2015.
2
The main point is always to antagonize the reductio of the phenomenon of tone to
the subjective alone, as Picardi (2007: 500) insists: “Tone is as much as conventional
and objective feature of word meaning as sense is, and Frege erred in confining it to
the realm of psychological association”.
3
It is in his “Retrospective Epilog” included in Grice 1989, that Grice speaks of
“Lower order” and “Higher order” Speech acts.
4
Thanks to Laura Caponetto for suggesting this connection.
5
Apparently we do not need a pejorative for an act of subordination, as in Langton’s
example “Blacks are not permitted to vote”.
6
Or, at least, sharing the presuppositions connected with the derogatory terms. Some
people may be unaware of the derogatory aspects of a term, either by not having
another “politically correct” term or just by not knowing the derogatory aspect of a
term in a context of a community. Travelling abroad may put people at risk of being
considered aither racist or simply unpolite just by ignorance.
7
I would like to thank Filippo Domaneschi for his suggestions on an early draft of
this paper and Paolo Leonardi for his careful reading and further suggestions, which,
unfortunately, I feel to have been unable to follow properly. A special thank to Laura
Caponetto for pointing out some mistakes and suggesting repair.

16