Content uploaded by Richmond Kwesi
Author content
All content in this area was uploaded by Richmond Kwesi on Nov 03, 2019
Content may be subject to copyright.
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
116
An Inferential Articulation of Metaphorical Assertions
Richmond Kwesi
University of Ghana, Department of Philosophy and Classics
rkwesi@ug.edu.gh
Abstract This paper argues for the view that metaphors are assertions by locating
metaphor within our social discursive practices of asserting and inferring. The literal and
the metaphorical differ not in the stating of facts nor in the representation of states of
affairs but in the kind of inferential involvements they have and the normative score-
keeping practices within which the inferential connections are articulated. This
inferentialist based account of metaphor is supplemented by insights from
accommodation theory. The account is significant for our understanding of both
metaphor‟s figurativeness and cognitive content.
Keywords: Assertion, Metaphor, Presupposition, Accommodation, Inferentialism
Received 11 January 2019; accepted 28 May 2019.
0. Introduction
The inferentialist-based approach to understanding metaphor pursued here is based on
Robert Brandom‟s (1994, 2000) inferential semantic account of discursive practices.
Brandom‟s inferentialism treats assertion from the point of view of the social practices
we engage in the game of „giving and asking for reasons‟, that is, it treats assertion on
the basis of the commitments and entitlements – the basic normative attitudes – that
speakers undertake when they make claims. This game of „giving and asking for reasons‟
implies that asserting and inferring are the fundamental kinds of doings in our social-
linguistic practices. The crucial argument I make is that, fundamentally, the making of
metaphorical claims can be understood from this game of „giving and asking for
reasons‟ – asserting and inferring – and it is on the basis of these things we do in using
metaphors that metaphorical claims have propositional contents that can be evaluated
for truth. However, Brandom‟s inferential account in its complexity presents an account
of literal meaning and content but metaphors have contents distinct from that possessed
by literal utterances. Hence, Brandom‟s account is modified and complemented with an
account of presuppositional accommodation. The modification is required to make the
normative social practices of „giving and asking for reasons‟ suitable for the
understanding and appreciation of metaphors. This modification is apt to show that
while both the literal and the metaphorical are indistinguishable in terms of what we
fundamentally do in using literal and metaphorical sentences, they differ mainly in the
kind of inferential involvements they have and the normative score-keeping practices
within which the inferential connections are articulated. The insight from
accommodation theory is to account for both the contextualism associated with
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
117
metaphors, and the acceptability of metaphorical moves in the language game despite
their literal impropriety. This inferential articulation of metaphor is significant in at least
two ways: one, it shows that contrary to Davidson (1979), Rorty (1987), Cooper (1986),
Reimer (2001), Lepore & Stone (2010, 2015) and others, metaphors can be regarded as
truth-evaluable assertions whose contents are conferred by the socio-linguistic practices
we engage in; and two, we gain an understanding of how the two dimensions – the
seeing-as phenomenological and the assertional – both belong to the phenomenon of
metaphor as has been argued especially by Moran (1989), Gaut (1997), Camp (2006,
2017), and Taylor (2016).
1. Brandom’s inferential articulation of linguistic practices
It has been observed that most of our utterances and statements stand for, represent,
and are about things and states of affairs in the world. According to Brandom, one
explanatory strategy to capture this observation, representationalism, is to begin with an
understanding of representation, truth and reference, «and on that basis explain the
practical proprieties that govern language use and rational action» (Brandom 1994: 69).
But Brandom thinks that we cannot have a suitable notion of representation or truth in
advance of our thinking about the correct use of our linguistic expressions. For Brandom,
the representational dimension of propositional content is intelligible only in the context
of «linguistic social practices of communicating by giving and asking for reasons in the
form of claims» (ivi: 153). Rather than using the concepts of truth and reference as
primitive or basic semantic concepts to the understanding of other semantic concepts
like meaning and inference, Brandom provides an account of our linguistic practices
that takes asserting and inferring as fundamental and then he explains truth and
reference in terms of their expressive and inferential roles. He privileges inference over
truth and reference by adopting an explanatory strategy that understands the meaning of
linguistic expressions and the conferment of propositional content in terms of the role
they play in reasoning.
Inference, for Brandom, is a kind of doing. So Brandom reverses the order of
explanation by starting with an account of what one is doing in making a claim and then
seeks to elaborate from it an account of what is said and the propositional content of
what is said. He understands asserting something as «putting it in a form in which it can
both serve as and stand in need of reasons: a form in which it can serve as both premise
and conclusion in inferences» (Brandom 2000: 11). He explains propositional
contentfulness in terms of being „fit to serve both as a premise and as a conclusion‟ in
inferences. Instead of construing saying (thinking, believing) that such and such in terms
of its correspondence to states of affairs in the world, Brandom asks us to understand it
«in terms of a distinctive kind of knowing how or being able to do something» and the
relevant sort of doing here is understood by its «inferential articulation» (ivi: 17). He
explains further:
Saying or thinking that things are thus-and-so is undertaking a distinctive kind of
inferentially articulated commitment: putting it forward as a fit premise for further
inferences, that is, authorizing its use as such a premise, and undertaking responsibility
to entitle oneself to that commitment, to vindicate one‟s authority, under suitable
circumstances, paradigmatically by exhibiting it as the conclusion of an inference
from other such commitments to which one is or can become entitled (ivi: 11).
Brandom is motivated, in part, by finding distinctive features that make human beings
sapient beings who have the capacity to engage in discursive or concept-using practices.
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
118
That is, he is interested in knowing what is involved in discursive beings undertaking
certain commitments by their utterances, how their utterances come to be assessed as
appropriate or inappropriate, and how certain consequences and implications follow
from their utterances. His answer to the above inquiries is that the linguistic practices
humans engage in are norm-governed social practices, the performances of which are
subject to normative attitudes of their practitioners and the attribution of normative
statuses to the performers of this social linguistic practice. What differentiates discursive
beings and their practices from that of non-discursive animals, according to Brandom is
«to be the subject of normative attitudes, to be capable of acknowledging proprieties and
improprieties of conduct, to be able to treat a performance as correct or incorrect»
(Brandom 1994: 32).
The sociality of the discursive practice lies in the normative statuses – the undertaking
and attributing of commitments and their attendant undertaking of responsibility for
those commitments – and normative attitudes – acknowledging the propriety or
impropriety of performances – while the linguistic nature of the practice lies in the
institution of the speech act of assertion – the making of claims that can stand for, and
be in need of, reasons. The making of claims, asserting, is intimately connected to the
practice of inferring in such a way that for Brandom, «asserting cannot be understood
apart from inferring» (ivi: 158); for the making of a claim is connected to the
consequences that follow from that claim and the way in which that claim can be used
as a reason for another claim. The inferential involvements of a claim or the network of
inferential connections that a claim has to other claims, the propriety of the speaker‟s
commitment to the claim and the appropriateness of the inferences one can make from
that claim to others, suffice to determine the semantic content of that claim. The
appropriateness of the inferences largely depends on, and is often determined by, the
normative statuses and attitudes, the commitments and entitlements of the performers
in a conversation.
Brandom understands the linguistic practices of asserting and inferring and the
institution of normative attitudes and statuses within the framework of a game which is
governed by rules and where certain moves or performances of the game players are
deemed appropriate or inappropriate. In this game, the conversationalists are players
and score-keepers who undertake commitments in virtue of their own moves, and
attribute commitments to other players on the basis of the appropriateness of their
moves too. A move in this game authorizes other moves by the speaker and licenses
others to make other moves or attribute commitments to the speaker. The speaker‟s
move can be challenged, and when it is challenged, the speaker undertakes the
responsibility to provide reasons to justify and vindicate his assertion, or she could
retract the assertion if the circumstance demands it. The correctness and
appropriateness of moves in the game determine the kind of normative attitudes that
score-keepers adopt to the performances of players in the game.
Brandom‟s inferentialist account of the semantic content of sentences, in a nutshell, is a
pragmatic account that treats the uses of language in terms of a social normative
practice where the making of assertions and the inferential involvements of assertions
are basic and principal to the understanding of language use. These practices are situated
within a framework where the making of assertions involves the undertaking of
commitments, the attribution of commitments to others, the licensing of others to make
certain inferences from the assertion, and the responsibility to justify one‟s assertions
when challenged.
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
119
2. The case for metaphorical assertions
A metaphor is often characterized as a figurative expression used to make someone see
one thing in terms of another thing. According to Lamarque and Olsen, a metaphor, by
its essence, is not to «state that something is the case» but rather, its constitutive aim is «to
invite or encourage a hearer to think of, conceive of, reflect on, or imagine one thing
(state of affairs, idea, etc.) in terms associated with some other thing (state of affairs,
etc.) often of a quite different logical type» (Lamarque & Olsen 1994: 360). Building on
Davidson‟s (1979) thesis that metaphors have no propositional contents, Lepore and
Stone contend that «interlocutors use their metaphorical discourse not to assert and
deny propositions but to develop imagery and to pursue a shared understanding»
(Lepore & Stone 2010: 177). As mere invitations to explore comparisons, metaphors are
thought not to be in the business of making claims or assertions that can be evaluated
for truth.
However, metaphors go beyond mere invitations and the experience of comparing two
things (Kwesi, 2018). Speakers and hearers of metaphors do more than what invitors and
invitees do; the practices they engage in with metaphors go beyond the intention to
issue invitations and the undertaking of thinking of one thing as another thing. A
plausible explanation for this something more, other than, and in addition to invitations
and undertakings, is that the issue of truth is at play. It is because speakers often make
assertions by their utterances of metaphors that they exhibit certain commitments like
endorsing and justifying metaphors. We will consider examples from Shakespeare to
illustrate these points.
Example 1:
Menenius: The senators of Rome are this good belly,
And you the mutinous members; for, examine
Their counsels and their cares; digest things rightly
Touching the weal o‟ the common; you shall find
No public benefit which you receive
But it proceeds or comes from them to you,
And no way from yourselves. – What do you think,
You, the great toe of this assembly?
First citizen: I the great toe? why the great toe?
Menenius: For that, being one o‟ the lowest, basest, poorest,
Of this most wise rebellion, thou go‟st foremost:
Thou rascal, that art worst in blood to run,
Lead‟st first to win some vantage.
But make you ready your stiff bats and clubs:
Rome and her rats are at the point of battle;
The one side must have bale.
[Coriolanus, Act 1 Scene 1]
This example illustrates the kinds of commitments that speakers of metaphors bring to
bear in the use of metaphors. What Menenius is doing in the second speech is to justify
why the First Citizen is the great toe of the assembly. Menenius makes a claim in the
first speech, his audience demands an explanation or justification for the claim, and
Menenius in the second speech, explains the claim, or elaborates on the claim, or
justifies the claim. Surely, if speakers do not intend to say or describe something, if they
do not intend to state that such-and such is the case, then they will not be committed to
endorsing, justifying, and withdrawing their metaphors.
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
120
Example 2:
King Henry: A speaker is but a prater, a rhyme is but a ballad, a good leg will fall, a
straight back will stoop, a black beard will turn white, a curled pate will grow
bald, a fair face will wither, a full eye will wax hollow, but a good heart, Kate, is
the sun and the moon, or rather the sun and not the moon, for it shines bright
and never changes but keeps his course truly. If thou would have such a one,
take me.
[The Life of Henry the Fifth, Act 5 Sc 2]
King Henry is wooing Katharine. In one sentence he says that a good heart is both the
sun and the moon, he then withdraws or retracts it, and says later that a good heart is
the sun and not the moon. Withdrawal or retraction is possible when what is retracted
or withdrawn is considered as a claim or an assertion. King Henry‟s withdrawal of his
initial metaphorical utterance is independent of his intention to invite Katharine to
explore certain comparisons.
Example 3:
Ely: But there‟s a saying very old and true:
“If that you will France win,
Then with Scotland first begin”.
For once the eagle England being in prey,
To her unguarded nest the weasel Scot
Comes sneaking and so sucks her princely eggs,
Playing the mouse in absence of the cat,
To „tame and havoc more than she can eat.
Exeter: It follows, then, the cat must stay at home.
Yet that is but a crushed necessity,
Since we have locks to safeguard necessaries
And pretty traps to catch the petty thieves.
[The Life of Henry, Act 1 Scene 2]
This example illustrates that the hearer can do more than just think of, conceive of, or
imagine one thing in terms of another thing. The hearer, like Exeter, can draw certain
conclusions and make inferences from the speaker‟s metaphorical utterances. The
hearer can only do these things if he or she construes the utterances of the speaker to be
making truth-evaluable claims.
What these examples establish are that:
a) there are certain commitments that bind speakers when they speak
metaphorically, and these commitments cannot be accounted for satisfactorily when the
invitational norm is thought to constitute the making of metaphors. Invitations do not
naturally come with certain commitments like endorsing and justifying – these
commitments naturally go with the issuance of claims and assertions.
b) The things speakers do with metaphors are not incompatible with their inviting
others to do certain things; but they are also not derived from, nor dependent on, their
aims to invite hearers to do certain things. Hearers recognize that speakers of metaphors
intend to make assertions or claims by their use of metaphors and hence, their
responses and reactions to metaphors – that they can question, endorse, deny
metaphorical attributions – suggest that metaphorical utterances are no different from
literal utterances as far as their being claims are concerned.
1
1
Indeed many accounts such as that of Carston (2002), Recanati (2004) and those based on relevance
theory such as that of Wilson & Carston (2006) subsume metaphor under a general category of „loose
talk‟ and argue that there are no special cognitive and interpretive differences between the literal and the
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
121
How do we explain these points from the point of view of an inferential approach to
metaphors?
3. An inferential articulation of metaphors
I shall argue below that Brandom‟s account of the propositional content of sentences
and the making of assertions as sketched above, complemented by insights from
presuppositional accommodation, is appropriate for treating metaphorical utterances as being
propositionally contentful, and for taking metaphorical claims as assertions that can be
evaluated for truth. In other words, an inferentialist account of the use of language
provides us with a theoretical framework within which metaphors count as assertions
and can be appraised for truth. Inferentialism does not necessarily give priority of place
to literal content and literal assertion. Brandom‟s account helps us to understand our
social, normative and linguistic practices in using language – both literal and
metaphorical uses of language. In focusing on what we are doing in making claims or
saying something, his account is neutral as to whether what we are saying is literal or
metaphorical. That is, the distinctive and basic kinds of doing – asserting and inferring –
that Brandom identifies are not peculiar to literal uses of language. As the examples
above show, in both literal and figurative uses of language, we engage in the drawing of
inferences: we put forth claims that can serve as, and stand in need of reasons. In both
literal and metaphorical uses of language, we can justify the claims we put forth, and we
can challenge the claims that other interlocutors put forward in our conversational
practices. Similarly, in asserting metaphorical and literal statements, we undertake certain
commitments and license others to make inferences from those claims.
So, inferring here is the common denominator in our linguistic practices of making
literal and metaphorical claims. This implies that the propositional content of our claims
will have to be inferentially articulated, since what can serve as, and stand in need of,
reasons must be propositionally contentful. The basic argument here, therefore, is this:
P1 – Putting forward a claim is putting forward something that can serve as
a reason and stand in need of reasons.
P2 – What can serve as, and stand in need of reasons, is propositionally
contentful
C: Hence, putting forward a claim is putting forward something that is
propositionally contentful.
What is important about this argument is that it is neutral to whether the claims put
forward are literal or metaphorical. It shows that there is what we can term as the
principle of uniformity in terms of the basic kind of doing with respect to literal and
metaphorical uses of language. This principle indicates that asserting and inferring are
neither specific to, nor distinctive of, literal uses of language. And since, on an
inferentialist account, these pragmatic practices confer semantic contentfulness on
sentences, literal and metaphorical sentences come to have propositional contents in
fundamentally the same way. In addition, this uniformity principle nullifies the
bifurcation between descriptive and non-descriptive uses of language and the
explanatory work that the distinction is supposed to achieve. For, since inferentialism
metaphorical. While these pragmatic accounts focus on the comprehension and interpretation of
metaphorical utterances, the account pursued here provides a motivation for why metaphorical utterances
can be regarded as claims and assertions in the first place. And the thought is that, they are claims in just
the same way as literal utterances are regarded as claims.
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
122
considers inferring, rather than truth and reference, as basic and primary, descriptive
contentfulness is not a useful determinant for understanding our uses of language.
Despite the sameness of the fundamental action we do in using literal and metaphorical
sentences, it can be argued that the game of giving and asking for reasons is suitable for
only literal sentences. For the rule-governed nature of the game, the undertaking and
attributing of commitments and the score-keeping by players, the propriety of the
moves in the game and how such moves are assessed, and the consequences of the
moves in the game and how they are determined – all these make sense within a literal
understanding of the discursive practices of sapient beings. But I think that there is a
possibility of extending the details of Brandom‟s account to include uses of metaphorical
sentences, and hence, saving the uniformity principle. This possibility is what I will
explore below.
The explanation pursued here is to treat metaphorically speaking (or speaking metaphorically)
as a presupposition marker when it is explicitly used to prefix an assertion or as tacitly
presupposed when a speaker asserts a sentence that is identified as a metaphor. This
presupposition is pragmatic in the sense used by Stalnaker (1973, 1974, 2002) and
Soames (1982), in that it is speakers and not sentences that presuppose anything.
However, this metaphorically speaking presupposition is not a proposition that is already
accepted as part of the common ground of a conversation; and it is not a proposition
that is a consequence of a (metaphorical) sentence, the truth or falsity of which affects
the semantic value of the sentence. It is a presupposition in the sense of signalling how
the assertion is to be understood and interpreted, and therefore, the inferential
involvement of the assertion is contingent on acceptance of the presupposition. The
metaphorically speaking presupposition, in its non-explicit form, marks a ternary relation
between the sentence, the context in which it is used, and the speaker‟s intention in
using the sentence. In saying that „Richard is a lion‟, for instance, a speaker presupposes
that he is speaking metaphorically when he is cooperating in the discourse and obeying the
Gricean rules of conversation and that the context and the circumstances of the
utterance makes it appropriate for construing the utterance metaphorically.
I shall adapt Soames‟ (1982) definition of „utterance presupposition‟ in explaining this
distinctive kind of presupposition of speaking metaphorically. The presupposition of
speaking metaphorically can be captured in this way:
An utterance U presupposes that the speaker S is speaking metaphorically at t
iff one can reasonably infer from U that S accepts that he is speaking
metaphorically and regards it as uncontroversial, either because
a. S thinks that the conversational context at t makes it appropriate to
construe U only as a metaphor, or because
b. S thinks that the audience is prepared to add the presupposition,
without objection, to the context against which U is evaluated.
2
Consider a simple conversation between two people about their head of department,
Schneider:
Tom: What do you think of this new head of department?
Dick: Schneider is a fox
2
This definition itself is not distinctive of metaphorically speaking. Indeed, it captures the general mode of
figuratively speaking, and as such can be extended to other figures of speech like irony and metonymy. What
will set metaphor apart from these other figures will be the kind of inferential connections speaking
metaphorically warrant.
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
123
Tom: Well, I think he is more of a serpent
Dick: Either way, he is treacherous.
Dick‟s initial utterance presupposes that he is speaking metaphorically and this
presupposition seems uncontroversial from the context in which he is talking about his
head of department. In this case, the fact that the conversation is about a human being,
coupled with the obvious literal falsity of his assertion, make it appropriate to construe
his assertion as a metaphor. The conversational context in this case determines that the
utterance be construed as a metaphor in virtue of the speaker‟s intention to speak
metaphorically and his interlocutor recognizing that he is speaking metaphorically.
Tom‟s recognition of Dick‟s speaking metaphorically informs his own use of a related
metaphor. The presupposition is sustained by Tom‟s recognition that Dick is
cooperating in the conversational discourse and expects that his utterance will be
construed metaphorically.
What is the purpose of this presupposition marker (i.e. metaphorically speaking) in the
score-keeping game of giving and asking for reasons? Among the moves the linguistic
game-player can make is the asserting of a metaphorical claim. In making this move, the
speaker presupposes that he is speaking metaphorically in the sense explained above.
The metaphorical claim itself usually will be literally inappropriate – it could be literally
false, semantically anomalous, pragmatically a misfire, a category mistake, and so on –
but the recognition of the presupposition (tacitly or explicitly expressed)
3
and the
contextual parameters within which the utterance is made leads to both the acceptability
and appropriateness of the move. Understanding the three key words italicized in the
preceding sentence – recognition of the presupposition marker, the appropriateness and
acceptability of the metaphorical utterance – and the relationship that exist among the
three, will shed light on the inferential articulation of metaphors within this Brandomian
framework.
A speaker‟s metaphorical utterance comes with what we have called the presupposition
of speaking metaphorically. The speaker, in the first place, relies on his/her hearer‟s
capacity to acknowledge this presupposition behind his utterance, and in the second
place, the speaker expects his/her hearer to recognize the presuppositional intent of
his/her utterance. Hornsby (1994) and Hornsby & Langton (1998) have characterized a
phenomenon that exists between users of language whereby the speaker and hearer
depend on a „mutual capacity for uptake‟ as users of language and a reliance on a
„minimal receptiveness‟ of users in their role as hearers, as reciprocity. According to
Hornsby and Langton, «people who share a language have the capacity not simply to
understand one another‟s words, but also to grasp what illocutionary acts others might
be trying to make» (Hornsby & Langton 1998: 25) and that in a successful linguistic
exchange «a speaker tries to do an illocutionary thing; a hearer recognizing that the
speaker is trying to do that thing is then sufficient for the speaker to actually do it»
(Hornsby & Langton 1998: 25). Reciprocity then allows that language users exploit the
capacity of others recognizing their illocutionary intent, like staking a claim, and the
capacity to acknowledge and receive the communicative intentions of others. One
3
When the speaker does not explicitly indicate that she is speaking metaphorically, usually, the features of
the sentence such as literal falsity or absurdity in addition to the context in which the sentence is uttered
can lead the hearer to recognize that the speaker was speaking metaphorically. This does not mean that
metaphors are necessarily identified as being literally false or semantically anomalous – the case of twice-
true metaphors present counter-examples to the view that metaphors are literally false. The identification
of a sentence as a metaphor is different from the essence or constitution of a metaphor. Literal falsity is
an identificational, rather than essential, feature of metaphor.
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
124
implication of this view is that the staking of a claim is not necessarily achieved by
convention or by its adherence to certain rules but by the audiencs awareness of the
speaker‟s intention.
The notion of reciprocity is apt here in grounding the linguistic moves that speakers
make which come with the metaphorical presuppositions. The recognition of the
presupposition marker on the part of the hearer and the speaker‟s expectation that the
hearer recognizes his/her presuppositions is dependent on this mutual capacity for
uptake and receptiveness. Reciprocity then suffices for regarding the metaphorical
utterances of the game-players as the making of claims, for their status as claims,
depends on whether the game players take them to be so. That is, by means of
reciprocity, the move of a game-player (when it is a metaphor) is successfully recorded
as the making of a claim by other game-players and score-keepers – and this is not
because the move is in accordance with the rules of the game per se but because it arises
from the point of view that speakers do something with their words and utterances.
Reciprocity alone, however, does not entail the acceptability of an utterance as a
metaphor: it needs to be supplemented by a process of presupposition accommodation to
account for the appropriateness and acceptability of metaphorical utterances in the
game. Accommodation, originally understood in relation to presuppositions refers to
the process or mechanism by which an utterance that requires a presupposition to be
acceptable, and where the presupposition was not part of the common ground of the
conversation before the utterance, that presupposition “comes into existence” (Lewis,
1979: 340) at the time of the utterance. That is, if the conversation requires that the
presupposition of an utterance is added to the conversational score for the utterance to
be acceptable, that presupposition is accommodated and added to the conversational
score. As a mechanism a hearer adopts to update the score of the conversation,
accommodation does not only imply that a hearer adjusts the context of the
conversation to receive the current utterance of the speaker but that the hearer also
makes a «tacit extension» (Karttunen 1974: 191) of the conversational context by
his/her acquiescence of the presupposition. By tacitly extending the context to
accommodate the speaker‟s presupposition, the hearer adopts a strategy that makes the
utterance of the speaker true and acceptable (Richard, 2004, 2008). The hearer updates
the conversation and the conversational score by adding the presuppositional
information to the shared common ground of the conversational context. The process
of accommodation, therefore, guarantees the appropriateness of the speaker‟s utterance,
for in deciding to acquiesce to the speaker‟s presuppositional suggestion, the hearer
comes to regard the speaker‟s utterance as an appropriate move in the game.
Roberts (2004: 511) identified two necessary conditions of presuppositional
accommodation, the satisfaction of which makes hearers accommodate, rather than
object to, the speaker‟s presuppositions:
a) Retrievability: what the hearer is to accommodate is easily inferable, so
that it is perfectly clear what is presupposed, and it is both salient and
relevant to the immediate context, and
b) Plausibility: the accommodated material leads to an interpretation that
is reasonable and unobjectionable in the context
In this distinctive kind of presupposition – metaphorically speaking – that comes with the
speaker‟s making of a metaphor, the process of accommodation ensures both the
acceptability and appropriateness of the metaphor. How so? When a metaphorical
utterance is made, an initial and perhaps, unreflective, reaction one can make is to see
the utterance as inappropriate – a wrong move in the conversational game. However, by
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
125
the hearer‟s recognition of the cooperative attitude of the speaker and the hearer‟s
recognition of the illocutionary intent of the speaker by means of reciprocity, the hearer
sets about to adjust the context and the conversation to admit the metaphorical
utterance. The acceptance of the metaphorical utterance, in turn, comes about by
inferring from the context that the speaker was speaking metaphorically. The hearer
then goes through the process of accommodation by accepting and adding the
presupposition to the conversational score and then updating the conversation to make
the metaphorical utterance of the speaker acceptable and appropriate. The initial
impropriety and awkwardness that seemingly greeted the metaphorical utterance goes
away; for, through accommodation, the metaphorical assertion becomes an appropriate
move in the game. Roberts‟ first condition of Retrievability is satisfied by the
recognition of the presupposition marker made possible by the context and the mutual
capacity of language users to recognize and be receptive to the illocutionary intent of
speakers. The second condition, Plausibility, is satisfied by the fact that it is only by the
accommodation of the presupposition that the meaning and interpretation of the
metaphor become contextually relevant and appropriate.
Accommodation also determines the kind of meaning we give to the metaphor and the
inferential involvements of the metaphor. It sets the inferential propriety and aptness of
a metaphor which accounts for the metaphor‟s acceptability. Accommodating a
presupposition associated with an utterance and updating the conversational score to
include the utterance implies that one understands the utterance. A speaker‟s preference
for uttering a metaphor in a conversation – when he/she could have chosen to speak
literally for the same effect – is borne out of the expectation that his/her hearer will
understand his/her utterance as metaphorical. And it is a linguistic fact that users of a
language do understand both the hackneyed and novel metaphors they use in their
linguistic exchanges. But what is it to understand an utterance – a metaphorical
utterance for that matter?
Understanding a metaphor may involve seeing one thing (Schneider) as another thing (a
fox). This is the phenomenological or figurative dimension to metaphor. Explicitly qualifying
one‟s assertion that one is speaking metaphorically or implicitly presupposing that one is
speaking metaphorically means that the assertion should not be construed in terms of
the normal signification of the words of the assertion. Recognizing the presupposition
and accepting the metaphorical utterance is the way in which accommodation embraces
the phenomenological dimension of the metaphor. Understanding a metaphor,
however, goes beyond the experiential seeing-as; it involves, among others, the ability to
reason with the metaphor, the ability to use the metaphor in other contexts, the ability
to draw inferences from the metaphor, and the ability to use the metaphor as a premise
or conclusion of an argument (Macagno & Zavatta 2014; Oswald & Rihs 2014;
Wagemans 2016; Ervas, Gola & Rossi 2018): this is the assertional dimension of a
metaphor. This is the sense of putting forward a metaphorical claim in the form that is
fit to serve as, and stand in need of reasons. In other words, this is the dimension of the
dual mode of asserting-and-inferring that is associated with metaphors. By
accommodating a metaphor, the hearer reflects his/her understanding of the metaphor,
where this understanding consists of both the phenomenological and assertional
dimensions of a metaphor. The relationship between the two dimensions is such that
the phenomenological dimension which accounts for the presupposition of
metaphorically speaking determines the kind of inferential involvement of the
metaphor, and hence, it determines the assertional dimension of the metaphor. The
correctness and appropriateness of the inferential involvements of a metaphor are set
and constrained by the phenomenological dimension of the metaphor through the
process of accommodation.
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
126
The role of accommodation in grounding both the assertional and phenomenological
dimensions of metaphor stems from the idea of accommodation as the making of a tacit
extension to the context of conversation or as the adjusting of the extension of a term
in order to make an utterance acceptable and true. Richard (2004, 2008) used the
notions of accommodation and „contextual negotiation‟ to argue for the truth relativity
of expressions whose meanings depend on the context in which they are used. Richard
explained that it is the processes of accommodation and negotiation that accounts for
why a sentence like „Mary is rich‟ expresses a claim whose truth is relative. The
significant insight from Richard here is that in making an assertion that „Mary is rich‟
one is simultaneously inviting others to conceptualize the way in which Mary „counts as‟
being rich. Mary may count as rich in the pool of university professors but not rich in
the pool of billionaires like Bill Gates. A similar insight is what obtains in the making of
metaphorical assertions which corresponds to the phenomenological and assertional
dimensions. In saying that Schneider is a fox, Dick is making a claim and simultaneously
inviting Tom to think of the way in which Schneider counts as a fox. By
accommodating the claim, Tom recognizes that Dick was speaking metaphorically by
means of the invitation to think of Schneider as a fox, and also that Dick was making an
assertion whose content will be given by its inferential role.
Once the presupposition of metaphorically speaking is accommodated and the
conversational context and score are updated to make the metaphorical claim
appropriate and acceptable, the inferential involvement of the metaphorical claim and
the propriety of the inferences from the metaphorical claim can be given. The deductive
and material inferences we can make from a literal claim are constrained by the rule-
governed nature of literal claims and the compositional analysis that is brought to bear
on the understanding of the inferential involvements of literal claims. For instance, from
the claim that „Schneider is a fox‟ we can logically infer that „Schneider is an animal‟ and
the propriety of this inference depends on the compositional meaning of the claim and
the rules that govern the making of deductive and inductive inferences.
A metaphorical claim, however, is not fettered by compositionality (Kwesi, 2019) and its
appropriateness is not dependent on the literal rules of the game. The making of the
claim, the acceptability of the claim, and the appropriateness or otherwise of that claim
depends on the existence of the metaphorical presupposition marker, the phenomenon
of reciprocity and the process of accommodation, all of which are sensitive to the
contextual parameters in which the claim is made. The inferences that can be made
from the metaphor and the propriety of those inferences depend on what counts as
Schneider being a fox. For instance, while „Schneider is crafty‟ is a consequence of her
being a fox in the context of the conversation, „Schneider is hairy‟ is not. In other
words, in seeing Schneider as a fox we can infer that she is crafty but the consequence
that she is hairy is not plausible when one sees her as a fox in the context of discussing
Schneider.
Tom‟s recognition of the presupposition of Dick‟s claim and his illocutionary intent of
making an appropriate claim, coupled with his accommodating and updating the
context, influences his own use of the metaphorical claim that Schneider is, rather, a
serpent. When Dick finally asserts that either way Schneider is treacherous, he does not
nullify their so taking their earlier assertions as metaphors; he rather shifts the
conversational context back to the literal way of talking, and he expects that Tom will
adjust the conversation accordingly.
4
4
What makes a particular inference good or bad, what makes a good inference good and a bad inference
bad, and the rules governing the making of such inferences will require a more comprehensive
framework, one that is beyond the current scope of this paper.
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
127
The undertaking of certain normative commitments and the attribution of
commitments to others that participants engage in by means of their assertions are
social practices that are relevant and appropriate in the making of metaphorical claims
as well as literal claims. The admission or acceptance of metaphorical claims in the game
of giving and asking for reasons as explored above implies that the dual statuses of
commitments and entitlements that characterize social practices and the playing of
games are applicable when the assertions game players make are metaphorical ones.
Similarly, the determination of the semantic content of assertions by means of their
inferential roles in reasoning is no different when the assertions are metaphorical ones.
How the literal differs from the metaphorical is not because the metaphorical is not fit
to be instituted in the game of giving and asking for reasons, but that players of this
game adopt a different mechanism in determining the propriety of the moves in the
game, the propriety of the inferential involvements of those moves, and the ways in
which the conversational contexts and scores are adjusted and updated.
4. Implications and merits of the inferential articulation of metaphors
The kind of explanatory strategy pursued here is a non-reductive, uniform, and inferentialist-
based approach to understanding metaphor. It is non-reductive in the sense that it does
not explain and evaluate the metaphorical in terms of the literal; it is uniform because it
does not provide a sui generis kind of „metaphorical truth‟ or „metaphorical assertion‟; it is
an inferentialist-based approach in treating metaphor from a pragmatically articulated
point of view by understanding the questions of truth, content, and assertion, in relation
to metaphor, in terms of what we do in using metaphors. For Brandom, instead of
starting with a metaphysical account of truth such as the correspondence theory and
using that to account for beliefs and assertions which are construed to be
representations that can be true, he offers an approach of understanding truth
ascriptions in terms of the act of calling something true. In other words, the emphasis is
on a pragmatic construal of truth, the act of calling something true, rather than the
descriptive content of what is associated with what is called true; this is to say that in
calling something true one is praising or endorsing it rather than describing it.
Taking a claim to be true, then, is undertaking a sort of „normative stance or attitude‟
towards that claim, that is endorsing it or committing oneself to it. Endorsing a claim or
committing oneself to it, according to Brandom, is understood «in terms of the role the
endorsed claim plays in practical inference, both in first-person deliberation and in
third-person appraisal» (1988: 77). Truth, on Brandom‟s view, is seen not as a property
independent of our attitudes but it is understood in terms of „taking-true‟ or „treating-as-
true‟. He writes that «in calling something true one is doing something, rather than, or in
addition to, saying something. Instead of asking what property it is that we are describing
a belief or claim as having when we say that it is true, [we] ask about the practical
significance of the act we are performing in attributing that property» (1988: 77).
Undertaking a normative stance or commitment towards a claim, we have seen, is
understood as putting it forward as fit to serve as, and stand for, the premise and
conclusion of an argument. Hence, the ascription of truth to the literal and the
metaphorical occur in fundamentally the same way.
The emphasis on the uniformity principle is not to deny that the metaphorical is in an
important sense dependent on the literal. But we need to be clearer on what the
dependency relation entails and what the primacy of the literal involves. The primacy of
the literal is in respect of the production and interpretation of metaphor: words do not
acquire mystical meanings when they are used in metaphors; and the literal meanings of
the words aid in the interpretation of the metaphor – indeed the interpretation of the
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
128
metaphor is often done using literal language. But the primacy of the literal should not
be extended to the appraisal of the metaphor: we get nowhere by insisting that
metaphors are literally false, absurd or inconsistent; we will simply be appraising
metaphor in literal terms. The fact that A depends, or is parasitic, on B does not entail
that A should be reduced to B: dependency does not entail reduction. The dependency
relation that exists between the literal and the metaphorical should be understood as a
dependency-for. So, the question to be asked is that the metaphorical depends on the literal
for what? A plausible answer is that the metaphorical depends on the literal for its
explication, interpretation, paraphrasing. This means that the literal has a communicative
priority over the metaphorical, and this communicative priority does not extend over to
the evaluation of metaphor.
To understand metaphor qua metaphor, to understand metaphor in relation to truth and
assertion, however, we must begin with our practices of using language in general, and
our use of metaphors in particular, and then account for the notions of truth and
assertion in terms of our linguistic practices. The virtue of the approach pursued here is
that it does not reduce and explain metaphor in literal terms; it does not begin with a set
standard to evaluate metaphor; and most importantly, it explains content, truth and
assertion, without a prior demarcation of, and which is oblivious to the differences
between, the literal and the metaphorical.
An inferential articulation of the basic kind of doing in using metaphor provides us with
a comprehensive account of the roles of the interlocutors and the linguistic expression
in our communicative practices. There is what the speaker does, what the hearer does,
and what the linguistic expression itself effects in hearers: these three-doings, each
independently necessary but jointly sufficient for our understanding of the doings
associated with metaphor. An account that focuses on what the metaphor brings about
– nudging us to see something, directing our attention to see certain similarities,
provoking certain thoughts in us – is focusing on but one of the three-doings, and as
such, an adequate and satisfactory account cannot be propounded from that. The
Davidsonian account is guilty of this one-sided approach to the use of metaphor. An
invitational account is primarily focused on what the speaker does, and so are other
accounts that focus on the intentions of the speaker. Similarly, the focus on the
acquisition of beliefs and the effects of metaphor on the part of the hearer is inadequate
in itself to give a comprehensive account of metaphor. But from an inferentially
articulated kind of doing in terms of commitments and entitlements, we get a basic and
comprehensive picture of what we do in using metaphors. By putting forward a
metaphorical claim the speaker undertakes a commitment by endorsing the claim as fit
for a premise or conclusion in reasoning. Whether the speaker, in undertaking such a
commitment thereby invites, encourages, suggests or proposes something to the hearer is
explainable in terms of the inferences that the speaker licenses and the hearer is entitled
to draw. That the speaker can endorse a claim, and provide justification or warrant for
the claim when it is challenged, stems from the fact that the claim put forward is fit to
serve as the premise or conclusion of an argument. Understood this way, the hearer of a
metaphor is not just a passive recipient of an invitation; the call to action, the directive
to observe something, and the promise of seeing one thing as another thing, are rooted
in the practical ability of the hearer to determine whether the claim put forward by the
speaker is appropriate to serve as the premise or conclusion in reasoning. The
interpretation the hearer can offer to the metaphorical statement is sensitive to the
inferential relations that she can derive from both the linguistic and the non-linguistic
context of the claim put forward.
Moran (1989), Camp (2006, 2015), Taylor (2016) and others have argued that in
addition to a seeing-as framing-effect dimension to metaphor, there is a propositional or
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
129
an assertion dimension to metaphor. Carston (2010), Indurkhya (2016) and others have
models for the comprehension of the two dimensions to metaphor. The two
dimensions are meant to explain metaphor‟s figurativeness and cognitive content. But
what is the nature and relationship between the two dimensions? It is tempting, as
Camp (2006) for instance does, to explain the assertional dimension in terms of speaker-
meaning or some form of Gricean implicatures. The approach pursued here differs
from the others in the literature by locating the two dimensions within a broader
theoretical framework of what pertains in our general practices of using metaphors. The
intentions of the speaker, or speaker-meaning in general, does not explain the
assertional dimension to metaphor; rather the significance of speaker-meaning and the
intentions of the speaker are understood and explainable in terms of the basic things
interlocutors do – their commitments and entitlements – in the game of „giving and
asking for reasons‟. What makes a metaphor an assertion, and what provides a metaphor
with its truth-evaluable content, is determined by its inferential role in our discursive
practices. Similarly, the models of comprehension and the cognitive processes hearers
go through in their understanding of metaphor that account for both dimensions take as
their starting point the contextual and pragmatic features associated with metaphor. The
approach pursued here augments such models in showing that the mechanisms for the
production and cognition of metaphorical and literal sentences proceed on the same
lines – primarily because we engage in the same practices in our use of literal and
metaphorical sentences – but the inferential approach explains the differences between
the literal and the metaphorical in terms of their inferential roles.
5. Conclusion
Brandom‟s inferentialism – which subjects semantics to pragmatics by explaining the
semantic contentfulness of assertions by their practical role in reasoning – offers a
suitable avenue for understanding the making of metaphorical claims. In that,
metaphorical claims are contentful in virtue of the role they play as appropriate premises
and conclusions in reasoning and argumentation. The propriety of metaphorical claims
and the propriety of the roles they play in reasoning are determined by the phenomenon
of reciprocity and the accommodative processes that conversational participants go
through in adjusting the conversational context to satisfy the presuppositional
requirement that participants are speaking metaphorically. Brandom‟s inferentialist
deontic-score-keeping framework in which he characterizes assertion and inferring, is
therefore extended to cover metaphorical claims by incorporating the notion of
metaphorically speaking as a presupposition marker, Hornsby‟s notion of reciprocity,
and presuppositional accommodation. In view of this extension, we show the
plausibility of the uniformity principle that indicates that literal and metaphorical
contentfulness are determined in the same way – by what we do in reasoning – asserting
and inferring. One implication of this view is that metaphorical claims can be „treated as‟
or „taken to be‟ true claims in the sense in which Brandom treats the ascription of truth
to assertions. From this inferential articulation of metaphors, we gain an understanding
of the relationship between the seeing-as phenomenological and assertional dimensions
of metaphors.
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
130
References
Brandom, Robert (1988), «Pragmatism, phenomenalism, and truth talk», in Midwest
Studies in Philosophy, vol. 12. n. 1, pp. 75-93.
Brandom, Robert (1994), Making it explicit: Reasoning, representing and discursive commitments,
Harvard University Press, Cambridge, MA.
Brandom, Robert (2000), Articulating reasons: An introduction to inferentialism, Harvard
University Press, Cambridge, MA.
Camp, Elizabeth (2006), «Metaphor and that certain je ne sais quoi», in Philosophical
Studies, vol. 129, pp. 1-25.
Camp, Elizabeth (2017), «Why metaphors make good insults: perspectives,
presupposition, and pragmatics», in Philosophical Studies, vol. 174, n. 1, pp. 47-64.
Carston, Robyn (2002), Thoughts and utterances: The pragmatics of explicit communication,
Blackwell, Oxford.
Carston, Robyn (2010), «Metaphor: Ad hoc concepts, literal meaning and mental
images», in Proceedings of the Aristotelian Society, vol. 110, pp. 297-323.
Cooper, David (1986), Metaphor, Basil Blackwell, New York.
Davidson, Donald (1979), What metaphors mean, in S. Sacks, ed., On metaphor, The
University of Chicago Press, Chicago, pp. 29-45.
Ervas, Francesca, Gola, Elisabetta & Rossi, Maria Grazia (2018), Argumentation as a bridge
between metaphor and reasoning, in S. Oswald, T. Herman, & J. Jacquin, eds., Argumentation
and language: Linguistic, cognitive and discursive explorations, Springer, Berlin, pp. 153-170.
Gaut, Berys (1997), «Metaphor and the understanding of art», in Proceedings of the
Aristotelian Society, New Series, vol. 97, pp. 223-241.
Hornsby, Jennifer (1994), Illocution and its significance, in S. L. Tsohatzidis, ed., Foundations
of speech act theory: Philosophical and linguistic perspectives, Routledge, London, pp. 187-207.
Hornsby, Jennifer, & Langton, Rae (1998), Free speech and illocution, in Legal Theory, vol. 4,
pp. 21-37.
Indurkhya, Bipin (2016), Towards a model of metaphorical understanding, in E. Gola & F.
Ervas, eds., Metaphor and Communication, John Benjamins, Amsterdam, pp. 123-146.
Karttunen, Lauri (1974), «Presuppositions and linguistic context», in Theoretical
Linguistics, vol. 1, pp. 181-194.
Kwesi, Richmond (2018), Metaphor, truth, and representation, in P. Stalmaszczyk, ed. Objects
of Inquiry in Philosophy of Language and Linguistics, Peter Lang, Berlin, pp. 117-146.
Kwesi, Richmond (2019 forthcoming), «Semantic meaning and content: The intractability
of metaphor», in Studia Semiotyczne.
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
131
Lamarque, Peter, & Olsen, Stein (1994), Truth, fiction, and literature: A philosophical
perspective, Clarendon Press, Oxford.
Lepore, Ernie & Stone, Matthew (2010), «Against metaphorical meaning», in Topoi, vol.
29, pp. 165-180.
Lepore, Ernie & Stone, Matthew (2015), Imagination and convention: Distinguishing grammar
and inference in language, Oxford University Press, Oxford.
Lewis, David (1979), «Scorekeeping in a language game», in Journal of Philosophical Logic,
vol. 8, pp. 339-359.
Macagno, Fabrizio & Zavatta, Benedetta (2014), «Reconstructing metaphorical
meaning», in Argumentation, vol. 28, n. 4, pp. 453-488.
Moran, Richard (1989), «Seeing and believing: Metaphor, image and force», in Critical
Inquiry, vol. 16, n. 1, pp. 87-112.
Oswald, Steve & Rihs, Alain (2014), «Metaphor as argument: rhetorical and epistemic
advantages of extended metaphors», in Argumentation, vol. 28, n. 2, pp. 133-159.
Recanati, Francois (2004), Literal Meaning, Cambridge University Press, Cambridge.
Reimer, Marga (2001), Davidson on metaphor, in P.A. French & H.K. Wettstein, eds.,
Midwest studies in philosophy, 25, Figurative language, Blackwell, Boston (Mass), Oxford
(UK), pp. 142-55.
Richard, Mark (2004), «Contextualism and relativism», in Philosophical Studies: An
International Journal for Philosophy in the Analytic Tradition, vol. 119, n. 1, pp. 215-242.
Richard, Mark (2008), When truth gives out, Oxford University Press, Oxford.
Roberts, Craige (2004), Pronouns and definites, in M. Reimer & A. Bezuidenhout, eds.,
Descriptions and beyond, Oxford University Press, Oxford, pp. 503-543.
Rorty, Richard (1987), «Unfamiliar noises I: Hesse and Davidson on metaphor», in
Proceedings of the Aristotelian Society, vol. 61, pp. 283-311.
Soames, Soames (1982), «How presuppositions are inherited: A solution to the
projection problem», in Linguistic Inquiry, vol. 13, n. 3, pp. 483-545.
Stalnaker, Robert (1973), «Presuppositions», in Journal of Philosophical Logic, vol. 2, n. 4,
pp. 447-457.
Stalnaker, Robert (1974), Pragmatic presuppositions, in M. Munitz & P. Unger, eds.,
Semantics and Philosophy, New York University Press, New York, pp. 197-214.
Stalnaker, Robert (1978), Assertion, in P. Cole & J. Morgan, eds., Syntax and Semantics,
Academic Press, New York.
Stalnaker, Robert (2002), «Common ground», in Linguistics and Philosophy, vol. 25, n. 5,
pp. 701-721.
RIFL (2019) Vol. 13, n. 1: 116-132
DOI: 10.4396/09201905
__________________________________________________________________________________
132
Taylor, Charles (2016), The language animal: The full shape of the human linguistic capacity,
Harvard University Press.
Wagemans, Jean H. M. (2016), «Analyzing metaphor in argumentative discourse», in
RIFL – Rivista Italiana di Filosofia del Linguaggio, vol. 102, pp. 79-94.
Wilson, Deirdre & Carston Robyn (2006), «Metaphor, relevance and the „emergent
property‟ issue», in Mind & Language, vol. 21, n. 3, pp. 404-433.
