Microsoft Word - mckeonm,+Inference+claims+as+assertions_PROOFS.doc
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
Inference Claims as Assertions
MATTHEW W. MCKEON
Michigan State University
South Kedzie Hall
East Lansing, MI 48824-1032
mckeonm@msu.edu
Abstract: When a speaker states an
argument in arguing—in its core
sense—for the conclusion, the speak-
er asserts, as opposed to merely
implies or implicates, the associated
inference claim to the effect that the
conclusion follows from the premises.
In defense of this, I argue that how an
inference claim is conveyed when
stating an argument is constrained by
constitutive and normative conditions
for core cases of the speech of arguing
for a conclusion. The speech act of
assertion better reflects such condi-
tions than does implication, conversa-
tional implicature, or conventional
implicature.
Résumé: Lorsqu'un locuteur avance
des raisons pour appuyer une conclu-
sion, dans le sens fondamental
d’argument, il affirme, et non sim-
plement implique ou insinue,
l’inférence associée que la conclusion
découle des prémisses. Pour défendre
ceci, je soutiens que la façon dont une
affirmation d'inférence est exprimée
lorsqu’on avance un argument est
limitée par des conditions constitu-
tives et normatives pour les cas
fondamentaux du discours d'argumen-
tation. L'acte de parole de l'assertion
reflète mieux ces conditions que
l'implication, l'implicature conversa-
tionnelle ou l'implicature convention-
nelle.
Keywords: arguments, arguing for a conclusion, assertion, inference claims,
implicature
1. Introduction
In arguing for a conclusion by stating an argument for it one con-
veys an inference claim to the effect that the conclusion follows
from the premises. How is the inference claim conveyed by stating
the argument? Is it merely implied, implicated in some way, or
asserted? In this paper, I develop a response in two steps. First, I
clarify what I take to be arguing for a conclusion in its core sense
in a way that highlights the significance of inference claims. Sec-
ond, based on this clarification, I argue that by stating an argument
360 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
in arguing for its conclusion one asserts the associated inference
claim as opposed to merely implying or implicating it.
More specifically, I first argue in favor of the following three
claims. (i) One isn’t arguing for a conclusion in its core sense
unless one believes the associated inference claim. (ii) Arguing for
the conclusion in this way doesn’t pass normative muster unless
the inference claim is true or acceptable. (iii) When so arguing for
the conclusion the inference claim that is expressed is a primary
point in stating the argument. I then argue that given (i)-(iii) when
a speaker S states an argument in core cases of arguing for its
conclusion S asserts the associated inference claim and doesn’t
merely imply or implicate it.
It has been maintained in the literature that what I am calling
inference claims are implied by what statements of arguments
express (Hitchcock 2011, p. 214), conveyed as conversational
implicatures (Bermejo-Luque 2011a, p. 64) conveyed as conven-
tional implicature (Grice 1989, pp. 25-26), and implicitly asserted
when stating arguments (Hitchcock 2011, pp. 191-192). I believe
that this paper is the first to justify that they are asserted by appeal-
ing to (i)-(iii), which I take to be motivated by the notion of argu-
ing for a conclusion in its core sense. With respect to such core
cases of arguing for a conclusion, your statement of an argument
does not implicate the corresponding inference claim and what you
express only implies it in the trivial sense that what you assert you
thereby imply.
I start by clarifying the notion of an argument in its reason-
giving sense, distinguishing between such arguments and state-
ments of them. I then introduce the concept of an inference claim,
which a speaker S conveys by her statement of an argument in its
reason-giving sense. The question arises how S expresses an infer-
ence claim by making such a statement. In response, I first clarify
what I take to be core cases of arguing for a conclusion. I then
argue that S’s statement of an argument counts as arguing for the
conclusion in such ways only if S asserts the associated inference
claim in stating the argument. Finally, I conclude.
Inference Claims as Assertions 361
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
2. Preliminaries: arguments, statements of arguments, and
inference claims
2.1 Arguments and statement of arguments
Two senses of “argument” in the informal-logic literature are the
formal-logic and argumentation senses. According to the first, an
argument is an ordered pair. The first member is the set of premis-
es, the second is the conclusion. An argument in the argumentation
sense is a complex speech act in which an arguer presents a thesis
to an audience and defends it with premises advanced as reasons
for the thesis. Goldman (1994, p. 27) calls the formal logic and
argumentation senses of “argument,” the abstract and social sens-
es, respectively. Biro and Siegel (2006, p. 92) speak of a similar
distinction between arguments as abstract objects and arguments
in use. As Goldman suggests, these senses needn’t be competing.
An argument in the argumentation sense advances the premises of
an argument in the formal-logic sense as reasons for the conclu-
sion, i.e., the thesis the arguer aims to defend.
In this paper, both the abstract and social senses of argument
are in play. Specifically, I take an argument to be a premise-
conclusion complex of propositions. For simplicity of exposition,
I’ll focus on simple (as opposed to extended) arguments and so at
the start take an argument to be an ordered pair of a set of proposi-
tions (premises) and a conclusion proposition. Borrowing from
Hitchcock (2007, p. 103), I take the reason-giving sense of argu-
ment to be an initial conception of argument as a type of discourse
in which the arguer expresses a point of view (encapsulated in the
conclusion) and offers one or more reasons (the premise(s)) in
support of the conclusion. The reason-giving sense of argument is
an initial conception of a use of an argument qua premise-
conclusion complex of propositions according to which the prem-
ises are advanced, collectively, as reasons in support of the con-
clusion. The reason-giving sense of argument is akin to Goldman’s
social sense of argument: an argument in the social sense is an
argument in the reason-giving sense.
Arguments are used in reason-giving ways to realize a variety
of aims such as: (i) to communicate one’s reasons for which one
accepts the conclusion (Thompson 1967); (ii) to communicate
362 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
hypothetical reasoning (Meiland 1989); and (iii) to persuade an
addressee of the truth of the conclusion (Pinto 2001).
(i) A: Where’s Beth?
B: Beth is listening to music on the stereo and so is in
the living room.
(ii) A: Where’s Beth?
B: If Beth is listening to music on the stereo and so is in
the living room, then she doesn’t want to be disturbed.
(iii) A: Beth is not in the house.
B: She is. She is listening to music on the stereo.
In exchanges (i)-(iii), B’s response uses an argument in its reason-
giving sense in order to communicate her reason for which she
believes the conclusion, to communicate what B takes to follow
from a supposition, and to persuade A of the conclusion, respec-
tively. B’s responses state argument C.1
Beth is listening to music on the stereo
\ Beth is in the living room
By ‘state’ I mean ‘express.’ If one asserts something or merely
gives voice to it, then one expresses and so states it. To state an
argument is to express it. There is a process/product ambiguity
with respect to “the statement of an argument.” For example, “the
statement of C” designates both the speech act of stating argument
C and what is stated. B’s statement of C in (i) refers to B’s speech
act of stating C and the proposition stated, i.e., Beth is listening to
music on the stereo and so is in the living room. In what follows,
I’ll let the sentential context disambiguate my uses of “statement
of an argument.”
No statement of an argument is equivalent with the argument
that is stated. One commonly acknowledged rationale is that an
argument can be stated in different ways (e.g., see Vorobej 2006,
1 This is to read B’s response in (ii) as stating argument C in the antecedent.
Inference Claims as Assertions 363
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
p. 9). For example, the following two sentences can be used to
express one and the same argument: “Not every Republican is
rich, so some Republicans won’t benefit from the tax cuts”; “some
Republicans are not rich; consequently, not every Republican will
benefit from the tax cuts.” These sentences so used are different
statements of one and the same argument.
2.2 Inference claims
In contrast to the formal logician’s use of arguments in the abstract
sense, arguments in the reason-giving sense are typically charac-
terized as arguments (in the abstract sense) used to claim implicit-
ly or otherwise that some propositions support the truth of another.
Often, informal-logic texts characterize arguments in their reason-
giving sense and associate them with claims to the effect that some
propositions support another proposition by virtue of being evi-
dence, grounds, or reasons for its truth (e.g., Kelley, 2014, p. 69;
Copi and Cohen 2005, p. 6; and Govier 2010, p. 3).
With respect to the reason-giving sense of argument, Hitchcock
remarks that, “[w]hat is crucial to an argument is the claim that the
reasons offered collectively support the conclusion” (2007, p.
105). Hitchcock calls such a claim an “inference claim.” It is
commonly held in the argumentation and informal logic literature
that a claim to this effect is essential to an argument in its reason-
giving sense. E.g., see Scriven 1976, p. 84; Grennan 1994, p. 187;
Groarke 2002, p. 51; Vorobej 2006, pp. 8-9; and Bermejo-Luque
2011a, p. 90.
We may think that there are two independent roles for “there-
fore” and its cognates: mere conclusion designator (e.g., Corcoran
1993)2; and the expression of a relation of support between prem-
ises and conclusion (e.g., see Epstein 2002, Hamblin 1970, and
Hitchcock 2007, et al). The therefore of informal logic differs
from the formal-logic therefore, because informal logic considers
arguments in their reason-giving sense. In stating p therefore q in
2 Corcoran (1993, p. xx) emphasizes that an argument in this sense doesn’t per
se purport to show anything, and so it’s misleading to say that the conclusion is
based on the argument’s premise set or that the premises are given for the
conclusion. Corcoran remarks that his subsequent use of “\” to express argu-
ments is best not read as therefore but simply as conclusion.
364 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
order to advance p as a reason for q a speaker conveys that p is
reason on behalf of q (for textbook explications of this use of
therefore see Govier 2010, p. 3 and Fogelin and Sinnott-
Armstrong 2010, p. 51). In this way, the use of a therefore to state
an argument in its reason-giving sense is different from its use as
mere conclusion-designator in the presentation of an argument in
the abstract sense. This suggests that one’s statement of an argu-
ment is truth evaluable only relative to a reason-giving use of it.
On my view, an inference claim is crucial to an argument in the
reason-giving sense in two ways. First, a speaker S uses an argu-
ment to advance its premises as reasons that collectively support
the conclusion only if S claims implicitly or otherwise that the
conclusion follows—in some sense—from the premises. So,
claiming that the conclusion follows from the premises is constitu-
tive of using an argument in a reason-giving way. Second, S’s
reason-giving use of an argument doesn’t pass normative muster
unless the associated inference claim is true. If false, then S’s
attempt to establish the truth of the conclusion by advancing the
premises as reasons in support of it is unsuccessful. These two
connections between inference claims and reason-giving uses of
argument motivate thinking that the truth of an inference claim
bears on that of the statement of the corresponding argument. For
example, in conversational exchange (ii) above, B’s statement of
argument C is intuitively untrue if Beth being in the living room
doesn’t follow from her listening to music on the stereo.
In sum, the operative notion of argument in this paper is a
premise-conclusion complex of propositions that is distinct from
any statement of it. The statement of an argument used in a reason-
giving way is truth-evaluable and conveys an associated inference
claim. In what follows, I’ll adopt Hitchcock’s (2011) first-step
characterization of inference claims according to which an infer-
ence claim says, in effect, that the conclusion follows from the
premises.3
3 Hitchcock (2011) advances an analysis of follows from towards providing an
account of the general form of inference claims. My intent here in sticking with
this first-step characterization of inference claims is to remain neutral on its
theoretical development, which doesn’t have a direct bearing on the aims of the
paper.
Inference Claims as Assertions 365
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
How, exactly, is the inference claim so construed conveyed?
For example, when a speaker S utters, “p so, q” in order to ad-
vance p as a reason for the truth of q, how is the associated infer-
ence claim, q follows from p, conveyed by S’s statement of the
argument? Specifically, does S merely imply, implicate in some
way, or assert it? I assume that these alternatives are mutually
exclusive. The remainder of the paper is devoted to developing my
response.
Towards this end, I consider statements of arguments made in
core cases of arguing for their conclusions. I maintain that if a
speaker S’s statement of an argument counts as a core case of
arguing for its conclusion, then S conveys the corresponding infer-
ence claim by means of asserting it. In the next section, I clarify
the paper’s operative sense of arguing for a conclusion in a way
that motivates constraints on accounts of how inference claims are
conveyed. In section 4, I deploy these constraints to make the case
that one asserts the inference claim when stating an argument in
arguing for its conclusion as opposed to merely implying or impli-
cating it. Here I remain neutral on the manner inference claims are
conveyed when reason-giving arguments are stated by speakers
who are not arguing for their conclusions in the sense operative
here. So, I only partially answer the question posed one paragraph
back.
3. The complex speech act of arguing for a conclusion
In order to clarify the paper’s operative sense of arguing for a
conclusion, I highlight three facets of the connection between
inference claims and this complex speech act. First, stating an
argument doesn’t count as arguing for its conclusion in its core
sense unless the arguer believes the associated inference claim.
Second, arguing for its conclusion in this way is normatively
problematic unless the inference claim is true and the arguer is
justified in believing the inference claim. Third, the inference
claim conveyed by one’s statement of an argument is not informa-
tionally and conversationally subordinate to the main content of
the statement of the argument.
366 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
3.1 Inference claims and argumentation
I take arguing for a proposition to be a form of what van Eemeren
and Grootendorst characterize as argumentation.
Argumentation is a verbal, social, and rational activity aimed at
convincing a reasonable critic of the acceptability of a standpoint
by putting forward a constellation of propositions justifying or re-
futing the proposition expressed by the standpoint. (2004, p. 1)
Arguing for a proposition, construed along the lines of this notion
of argumentation, is a complex speech act (1984, pp. 29-35). That
is, arguing for a proposition (pro-argumentation as opposed to
contra-argumentation, i.e., arguing against a proposition (1984, p.
9)) is reflected in a variety of argumentative moves that are speech
acts such as stating an argument, arguing against a proposed refu-
tation of an argument, and clarifying terminology (van Eemeren
and Houtlosser 2003, p. 87; van Eemeren and Grootendorst 2004,
p. 52 and 62, et al).
Drawing on van Eemeren and Grootendorst, I’ll consider the
speech act of arguing for a conclusion in the context of a critical
discussion, which I’ll construe simply as a discussion between a
protagonist (Pro) and an antagonist (Con) triggered by their desire
to resolve their difference of opinion with respect to some claim
(the standpoint).4 The role of Pro in the discussion is to defend the
standpoint; Con’s role is to raise doubts about it (2004, pp. 60-61).
Their aim is to resolve their difference of opinion. It’s realized
only when Pro and Con have reached agreement on the acceptabil-
ity of the standpoint (2004, pp. 57-58).
Suppose Pro argues for the standpoint by stating an argument in
its reason-giving sense, i.e., by advancing premises as reasons in
support of the standpoint represented by the conclusion. What
4 Van Eemeren and Grootendorst provide a theoretical model of the ontology of
a critical discussion (2004, Ch. 3), as well as operative norms (2004, Ch. 8). A
starting point for their theorizing is their intuitive characterization of a discus-
sion between participants who aim to resolve a difference of opinion by ex-
changing arguments and advancing reasoned responses to criticisms (e.g., 1984,
pp. 1-2, where critical discussions are referred to as “argumentative discus-
sions”). For an alternative but related account of the speech act of arguing see
Bermejo-Luque (2011a, Ch. 3).
Inference Claims as Assertions 367
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
makes Pro’s speech act of stating an argument count as arguing for
the conclusion? Van Eemeren and Grootendorst contend that Pro
isn’t arguing for the conclusion unless Pro aims to persuade Con
of it (2004, p. 53; 1984, 47-49).5 In order to highlight a connection
between inference claims and arguing for a conclusion, I now
motivate another necessary condition.
Following Searle’s (1969) way of distinguishing types of
speech acts, van Eemeren and Grootendorst advance a number of
conditions constitutive of successfully arguing for a conclusion O
by adducing premise-statements S1, S2, ... , Sn (1984, pp. 39-45).
One such success condition is the sincerity condition.
S believes that the constellation of statements S1, S2 (, . . . , Sn)
constitutes an acceptable justification of O” (1984, p. 44).
A speaker is not successfully arguing for a conclusion O by adduc-
ing premises unless the speaker believes that the premises consti-
tute an acceptable justification of O.6 Granted that there are other
senses of “arguing for a conclusion,” I take the sincerity condition
to reflect a core sense of arguing for a conclusion by stating an
argument. By arguing for a conclusion in this sense, one advances
what may be labelled an a persona (from the person) argument,7
5 Van Eemeren and Grootendorst approvingly note (1984, p. 36) Searle’s
remark that “‘I am simply stating that p and not attempting to convince you’ is
acceptable and ‘I am arguing that p and not attempting to convince you’ is not”
(1969, 66). I take Searle’s remark to be compatible with acknowledging uses of
arguments other than as instruments of persuasion. E.g., to figure out what to
believe, to demonstrate that one has worked out an inference claim for a tutor.
This turns on the plausible claim that such uses of argument do not constitute
acts of arguing for their conclusions. I don’t have space to elaborate.
6 Van Eemeren and Grootendorst worry that the sincerity condition is too
speaker-centered and so doesn’t reflect the obligations of arguers when they
don’t believe that the premises they advance justify the conclusion. They think,
in effect, that if a respondent R justifiably takes a speaker S to be arguing for a
conclusion in a critical discussion, then R is entitled to hold S responsible for
the inference claim even if S doesn’t believe the inference claim (1984, p. 42).
This puts S on the hook to defend the inference claim if challenged by R (note
20 p. 195). As I discuss just below, I take the sincerity condition to enable an
intuitive doxastic requirement normative for acts of arguing for a conclusion.
7 I thank the reviewer for suggesting the notion of a persona argument to clarify
the paper’s operative sense of arguing for a conclusion.
368 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
which is a type of argument in its reason-giving sense. Such an
argument is put forward as the author’s own argument for the
conclusion. In using a persona argument to argue for its conclu-
sion, a speaker S advances the premises as S’s reasons to believe
the conclusion. Specifically, the premises of such an argument are
among S’s reasons for which S accepts the conclusion.
The author of such an argument believes the premises consti-
tute an acceptable justification of its conclusion, and so believes
that the conclusion follows from them in some way. This grounds
a cognitive connection between inference claims and arguing for a
conclusion that I take to be reflected by the sincerity condition.
Specifically, arguing for a conclusion in the operative sense here
requires that one believe the inference claim associated with the
corresponding a persona argument. If you state an argument
whose inference claim you do not believe, then you do not state an
a persona argument. Accordingly, you aren’t arguing for the
conclusion in the way that is of concern here. Hereafter, unless
specified otherwise, this is the operative notion of arguing for a
conclusion,8 and a persona argument is the corresponding type of
argument.
To summarize, in order for a speaker S’s statement of an argu-
ment to count as an act of arguing for the conclusion, S must
believe the associated inference claim. This is constitutive of the
act of arguing for a conclusion by stating the corresponding argu-
ment. This constitutive connection between inference claims and
arguing for a conclusion is reflected by Belief.
8 Clearly, this notion of arguing is not idiosyncratic. For example, according to
Bermejo-Luque (2011a) a constitutive goal of a speaker S’s arguing for a
conclusion is to show that a target claim is correct (p. 53), and the correspond-
ing sincerity condition requires, in effect, that S believes what I refer to as the
inference claim (p. 72). Also, Aiken and Talisse (2019, p. 11) take, as a starting
point, an argument advanced by a speaker S to be S’s attempt not only to make
clear what S’s reasons are for her conclusion-belief, but also to vindicate or
defend what S believes by showing that her conclusion-belief is well-supported
by compelling reasons. I take it that according to these notions, the premises S
advances for the conclusion are among the reasons for which S believes it. This
is the operative sense of arguing here.
Inference Claims as Assertions 369
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
(Belief) You express an inference claim by your statement of
an argument in arguing for its conclusion only if you believe
the inference claim.
If you don’t believe the inference claim, your statement of an
argument can’t count as an act of arguing for the conclusion be-
cause, by Belief, you don’t express the inference claim.9 Accord-
ingly, Belief grounds a pragmatic criterion that S’s statement of an
argument must satisfy in order to count as a successful act of
arguing for the conclusion. Specifically, S must believe the infer-
ence claim in order for her statement of the argument to express it.
3.2 Inference claims and argumentation norms
On my view, there are argument-centric and speaker-centric nor-
mative features of arguing for a conclusion. The argument-centric
features are the properties that the argument stated should have in
order for the statement of it to realize the aim(s) of the relevant
critical discussion. An argument’s possession of these properties
makes good its use to argue for the conclusion. One of these good-
making features F determines that the associated inference claim is
good.10 An argument’s possession of F is significant, in part,
because the statement of the argument advances the aims of the
relevant critical discussion only if the associated inference claim is
good.
This motivates a speaker-centric normative feature of arguing
for a conclusion. S’s use of an argument to argue for its conclusion
in a critical discussion passes normative muster only if S is justi-
fied in believing that the argument advances the aim(s) of the
discussion. Again, an argument advances the aims of critical dis-
cussion only if the associated inference claim is good. Therefore,
S’s use of an argument to argue for its conclusion in a critical
9 Although you could certainly be taken to express one. What it is to “count as
an act of arguing for the conclusion” is interestingly ambiguous in this way, just
like promising – between whether it is really that or just appears to be and so is
taken to be.
10 I don’t take rhetorical features such as being comprehensible to the intended
audience and on-topic as such features.
370 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
discussion passes normative muster only if S is justified in believ-
ing that the associated inference claim is good.
Clarification of what makes an inference claim good turns on
the identification of the corresponding good-making feature F of
arguments. Therefore, different accounts of F generate different
accounts of what makes inference claims good ones. To illustrate,
I’ll use the Objective Epistemic and Pragma-dialectic approaches
to argumentation norms. To be sure, neither approach features
inference claims. Nevertheless, I’ll borrow from these different
approaches to illustrate how the relevant good-making features of
arguments underwrite what makes inference claims good.
According to the Objective Epistemic Approach (OEA), the
primary aim of the rules of good argumentation is truth and the
avoidance of error (Goldman 1994, p. 29). An argument doesn’t
advance the aim of truth and avoidance of error unless its premises
justify the truth of the conclusion, i.e., unless the argument is
epistemically sound. An argument is epistemically sound only if
the corresponding inference claim is true. Therefore, an argument
advances the aim of truth and avoidance of error only if the corre-
sponding inference claim is true. So, the proper use of an argument
to argue for its conclusion (i.e., a use that advances the aim of
truth and avoidance of error) requires the user to be justified in
believing that the inference claim is true (e.g., see Goldman 1994,
p. 34, specifically argumentation rule 4c).
According to the Pragma-dialectic Approach (PDA), the prima-
ry aim of a critical discussion is rational consensus, i.e., the ration-
al resolution in a critical discussion of a conflict of opinion regard-
ing a standpoint (van Eemeren and Grootendorst 2004, p. 5). An
argument doesn’t advance the aim of such consensus unless it is
reasonable. If an argument is reasonable, then it’s both problem
valid and intersubjectively valid, i.e., it is relevant to resolving the
difference of opinion regarding the standpoint and it’s acceptable
to both Pro and Con (van Eemeren and Grootendorst 2004, pp. 16-
17). Therefore, the statement of an argument advances the aim of
consensus only if the corresponding inference claim is acceptable
to both Pro and Con. So, in order for a discussant’s use of an
argument to pass normative muster and thereby advance the aim of
consensus, she must be justified in believing that the inference
Inference Claims as Assertions 371
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
claim is acceptable to Pro and Con. Accepting the inference claim
is manifested in a commitment to defending it in light of criticism
and retracting it in case such criticism is unanswerable (van Eeme-
ren and Grootendorst 2004, pp. 54-55).
To summarize, the OEA and PDA inspired proposals for the
good making feature of arguments that make inference claims
good are epistemic soundness and reasonableness, respectively.
These different good-making features of arguments determine
different normative qualities of inference claims. According to the
OEA-inspired proposal, inference claims aren’t good unless they
are true. According to the PDA-inspired proposal, they aren’t good
unless they are acceptable to the involved parties. These different
proposals determine different speaker-centric normative features
of arguing for a conclusion. According to the OEA-inspired pro-
posal, a speaker S’s use of an argument to argue for its conclusion
is normatively successful only if S is justified in believing that the
corresponding inference claim is true. According to the PDA-
inspired proposal, this obtains only if S is justified in believing
that the inference claim is acceptable to all discussants.
That the sincerity condition is constitutive of successfully argu-
ing for a conclusion enables an intuitive, doxastic requirement that
is normative for acts of arguing for the conclusion. Specifically,
the sincerity condition enables the norm that in order for an act of
arguing for the conclusion to pass normative muster the speaker S
must be justified in believing the inference claim. If believing the
inference claim isn’t necessary to successfully argue for a conclu-
sion by stating an argument for it, why should it be an operative
norm that S’s believing the inference-claim be justified? What I
am pointing to here is that the sincerity condition has a norm-
activating function by virtue of determining a normative space
enabling doxastic norms that guide the speech act of arguing for a
conclusion by stating an argument.11
In short, a speaker S’s statement of an argument doesn’t count
as a good act of arguing for its conclusion unless the argument that
11 The sincerity condition isn’t determinative of which doxastic norms are
correct. On my view, the norms enabled by the sincerity condition are motivated
by the primary aims of the relevant dialectical context of the speech act. Differ-
ent accounts of these discursive aims may motivate different norms.
372 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
is stated is good and S’s is justified in believing this. An argument
is good only if the corresponding inference claim is good, i.e., is
true or acceptable. Therefore, S’s statement of an argument is a
good act of arguing for its conclusion only if (i) the inference
claim that is conveyed is true or acceptable, and (ii) S is justified
in believing that the inference claim is true or acceptable.
Intuitively, (1) the statement of an argument made in arguing
for its conclusion is true or acceptable only if the argument that is
stated is good. As the OEA and PDA inspired proposals illustrate,
(2) if the argument that is stated is good, then the associated infer-
ence claim is true or acceptable. Value is implied by (1) and (2).
(Value) If the statement of an argument made in arguing for
its conclusion is true or acceptable, then so too is the infer-
ence claim it conveys.
By the lights of Value, an argument being good (e.g., epistemically
sound, reasonable) bears on whether statements of them are true or
acceptable. Furthermore, if S is justified in believing that her
statement of an argument is true or acceptable, then Value enables
the transference of this justification to S’s belief that the inference
claim is also true or acceptable.12 Given that a statement of an
argument doesn’t count as a good act of arguing for its conclusion
unless (i) and (ii) (just above) obtain, Value underwrites the idea
that a speaker’s act of arguing for a conclusion by stating an ar-
gument is good only if the statement of the argument is true or
acceptable. This idea is intuitively plausible.
3.3 Inference claims and statements of arguments
Intuitively, when arguing for a conclusion a speaker expresses an
inference claim in stating an argument in order to realize a primary
aim of arguing for the conclusion. This motivates Primary.
12 Internalist accounts of justification may demand that S justifiably believes
Value in order for (ii) to obtain.
Inference Claims as Assertions 373
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
(Primary) Expressing the inference claim is a primary point
made by the statement of an argument in arguing for its con-
clusion, not a secondary one.
Primary reflects a connection between a primary point in stating
an argument and a primary aim in arguing for a conclusion. For
example, if speaker S’s primary aim in stating an argument in a
critical discussion is to persuade the respondent of the conclusion,
then the primary points conveyed by S’s statement of an argument
are what the addressee must accept in order to be persuaded by S’s
argument. The addressee must accept the inference claim in order
to be so persuaded. Hence, Primary. Alternatively, one may take
the primary aim of a complex act of argumentation to be showing
that a target claim is correct. If so, then the primacy (the non-
derivativeness) of expressing the inference claim is a consequence.
That is, if a primary aim of S’s in stating an argument in a critical
discussion is to justify acceptance of the conclusion to the ad-
dressees, then expressing the corresponding inference claim is a
primary point made by the statement of the argument.
4. How is an inference claim expressed by the statement of an
argument in arguing for its conclusion?
A question previously posed is: what makes Pro’s speech act of
stating an argument count as a speech act of arguing for the con-
clusion? In response, Pro’s speech act of stating an argument
counts as a speech act of arguing for the conclusion only if the
following three conditions obtain. First, Pro believes the corre-
sponding inference claim. Second, the truth of Pro’s statement
suffices for the truth of the inference claim that is conveyed. Third,
expressing the inference claim is a primary point of S’s statement
of the argument. More generally, the manner an inference claim is
expressed by the statement of an argument in arguing for its con-
clusion satisfies Belief, Value, and Primary. For ease of reference,
I repeat them so that they are in one place.
374 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
(Belief) You express an inference claim by your statement of
an argument in arguing for its conclusion only if you believe
the inference claim.
(Value) If the statement of an argument made in arguing for
its conclusion is true or acceptable, then so too is the infer-
ence claim it conveys.
(Primary) Expressing the inference claim is a primary point
made by the statement of an argument in arguing for its con-
clusion, not a secondary one.
Appealing to these conditions, I now make the case that if a speak-
er S argues for a conclusion by stating an argument, then S thereby
asserts the corresponding inference claim as opposed to merely
implying it, conversationally implicating it, or conventionally
implicating it. The case turns on the speech act of assertion fairing
best in satisfying Belief, Value, and Primary out of these alterna-
tives. The below scorecard summarizes the following discussion,
“N” and “Y” abbreviate “No and “Yes,” respectively.
(Belief) (Value) (Primary)
Mere Implication N Y Y
Conversational Implicature N N Y
Conventional Implicature N N N
Assertion Y Y Y
As the table indicates, Belief alone establishes that if a speaker S
argues for a conclusion by stating an argument, then S thereby
asserts the corresponding inference claim. Value and Primary are
worth discussing as a means of shedding further light on the status
of inference claims. I now justify the entries, starting with mere
implication.
4.1 Mere Implication
I label Mere implication as the position that an inference claim is
merely implied by what is expressed in stating the corresponding
Inference Claims as Assertions 375
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
argument in a critical discussion. Furthermore, I understand the
notion of implication semantically. Accordingly, I take the posi-
tion to hold that an inference claim is implied by virtue of the
meaning of one or more expressions that occur in the statement of
the argument.
For example, I take Hitchcock (2007, 2011) to hold that a
speaker’s statement of an argument in arguing for its conclusion
implies the corresponding inference claim. His rationale is three-
fold. First, an inference claim, “can be marked linguistically by
means of an illative expression (e.g., ‘therefore,’ ‘since’) govern-
ing the conclusion or reason” (2007, p. 106). Second, in order for a
statement of an argument to express an inference claim, it must
contain, explicitly or otherwise, an illative expression, which is in
sync with Hitchcock’s view that the components of arguments
include illative expressions (2007, pp. 106-107). Third, the infer-
ence claim is an analytical implication of the meaning of the illa-
tive expression contained (perhaps, implicitly) in the statement of
the argument. For example, “the word ‘so’ when used inferentially
implies, as part of its meaning and not as some pragmatic implica-
ture of its ordinary use, that the statement preceding it is relevant
to the statement following it… in the sense that it helps to estab-
lish the truth of the conclusion” (2011, p. 214).
Mere implication moves beyond Hitchcock in being explicit
that an inference claim is merely implied by the meaning of an
illative expression that occurs, implicitly or otherwise, in what is
expressed by the statement of an argument. If an inference claim
may be conveyed by the statement of an argument even though an
illative is not expressed, implicitly or otherwise, then this prob-
lematizes Mere implication. At any rate, Mere implication does
not satisfy Belief, because implication is simply a relation between
propositions that obtains regardless of a speaker’s beliefs.
That is, the logical connection doesn’t depend on your believ-
ing the implication. You say to me, “Beth is in the living room.”
“Beth is in the living room or in the kitchen” being an implication
of the proposition you state doesn’t depend on your believing the
implication. Furthermore, on my view, belief is not closed under
implication. Since what is implied by what you utter is independ-
376 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
ent of whether or not you believe what is implied, implication fails
Belief.13
Mere implication satisfies Value because, it preserves truth val-
ues. E.g., if p implies q and p is true, then q must be true. Also,
Mere implication seems to satisfy Primary. Consider, he is either
very smart or very silly, and he isn’t very smart uttered in order to
convey that he is very silly. This shows that you can intend the
implication of what you express. In stating an argument, you may
imply an inference claim in order to convey your belief of the
inference claim. However, implication can’t satisfy Belief this this
way, because, again, believing the implication of what you express
is not necessary for the implication to obtain.
4.2 Conversational implicature14
Does an arguer’s statement of an argument in a critical discussion
conversationally implicate the corresponding inference claim? An
affirmative response is derivable from Bermejo-Luque’s account
(2011a, 2011b) of how inference claims are communicated when
arguers argue for their conclusions by adducing premises in sup-
port of them. To illustrate, suppose that in the course of arguing
for a claim q in a critical discussion, Pro utters p so q in order to
persuade Con that q is true.
Following Bermejo-Luque (2011a, p. 64), Pro thereby implicit-
ly asserts the material conditional, if p then q, represented as p É
q. Bermejo-Luque thinks this is the inference claim (2011a, p.
13 Of course, this doesn’t rule out connections between the relation of implica-
tion and stating an argument. One might argue that stating an argument is
undermined if you aren’t prepared to at least defend the implications of that
statement.
14 My discussion of conversational and conventional implicatures aims to
present the elements of Grice’s view of implicatures relevant to answering the
question serving as section 4’s title. Towards this end, I draw from Grice’s 1967
publications “Logic and conversation” and “Further notes on logic and conver-
sation.” I rely on their reprints in Grice 1989. I also draw on the following
sources that critically survey Grice’s account of implicatures: well-known
textbooks on linguistic communication (Levinson 1983, Bach and Harnish
1979); handbook and encyclopedia articles (Horn 2004, Korta and Perry 2020);
review articles (Neale 1992, Potts 2007); and Bach’s oft-cited critical discus-
sions of Gricean implicatures in his 1994 and 1999 publications.
Inference Claims as Assertions 377
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
62).15 Bermejo-Luque contends that Pro’s intention to communi-
cate that q follows from p is conveyed as a conversational implica-
ture of Pro’s implicit assertion that p É q.16 By virtue of believing
that Pro abides by the Cooperative Principle and the conversation-
al maxims guiding the critical discussion, Con is entitled to think
that in asserting, p É q, Pro conveys as a conversational implica-
ture that q follows from p (2011a, p. 64 ; 2011b, p. 335).
Conversational implicatures are essentially calculable.17 Bor-
rowing from Bach and Harnish (1979, p. 169), Levinson (1983,
113-114), and Neale (1992, p. 527), I illustrate how Con’s reason-
ing might run in calculating the conversational implicature, q
follows from p, of Pro’s statement p so q.
(1) Pro says p so q and thereby implicitly asserts p É q;
15 Critics have pointed out that construing the content of inference claims in
terms of the material conditional drains the normative force from them (e.g.,
Pinto 2011, et al). In response Bermejo-Luque’s highlights the pragmatic aspect
of the normative force of inference claims (2011a p. 175; 2011b, p. 335). For
example, arguing that, Donald Trump likes asparagus, so 2 < 3, is problematic
(it’s hardly imaginable!), because even though Donald Trump likes asparagus É
2 < 3 is true by virtue of the truth of its consequent, it is not appropriate, by
Grice’s Maxim of Quantity, to assert it solely on the basis of knowledge of the
consequent. Hence, arguing as above is pragmatically flawed. However, Berme-
jo-Luque’s response doesn’t register why so arguing is epistemically flawed.
16 This requires that p É q be part of what is said in stating the argument. The
requirement is motivated by Grice’s view that a conversational implicature is
derived from what, if anything, is said by an utterance. On Grice’s strict concep-
tion of what is said, what is said by an utterance must correspond to “elements
of [the sentence uttered], their order, and their syntactic character” (1989, p. 87).
Following others (e.g., Bach 1999, p. 335), this does not mean that what is said
must be made fully explicit. For example, one can say that p É q is communi-
cated by a speaker’s utterance, p so q, by virtue of a correspondence between p
É q and “so.” This suffices for making p É q implicitly asserted and so (implic-
itly) stated by the statement of the corresponding argument. A fuller account
would spell out the nature of the correspondence between p É q and the conclu-
sion indicator.
17 “The presence of a conversational implicature must be capable of being
worked out; for even if it can be intuitively grasped, unless the intuition is
replaceable by an argument, the implicature (if present at all) will not count as a
conversational implicature” (Grice 1989, p. 31).
378 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
(2) There is no reason to suppose that Pro is not observing
the conversational maxims and cooperative principle
that guide the critical discussion;
(3) Only if Pro thinks that q follows from p is Pro’s saying
p É q consistent with the presumption that Pro is ob-
serving the maxims and cooperative principle;18
(4) Pro knows (and knows that I know that Pro knows)
that I can see that Pro thinks the supposition that Pro
thinks that q follows from p is required;
(5) Pro has done nothing to stop me thinking that q follows
from p.
So, from (1)-(5), (6) Pro intends me to think, or is at least willing
to allow me to think that q follows from p.
So, from (6), (7) Pro has implicated that q follows from p.
Steps 1-7 reflect three speech-act moves Pro makes in arguing
for q: (i) stating, p so q, thereby (ii) (implicitly) asserting p É q,
and (iii) conversationally implicating that q follows from p. Con’s
reasoning, embodied in 1-7, reflects Con’s belief that Pro’s (im-
plicit) assertion that p É q is not sufficiently informative in the
context of Pro’s arguing for q. So, the conveyance of q follows
from p requires Con’s inference from the supposition that Pro
wouldn’t have said p É q if Pro hadn’t meant something more than
that, i.e., hadn’t also meant q follows from p. I now make several
comments that inform my evaluation of conversational implicature
as the mechanism by which inference claims are expressed.
(a) Step 5 registers the cancelability of conversational im-
plicatures (following Grice 1989, p. 44). This makes
the pragmatic inferences that arise from such implica-
tures defeasible. In this way, they are more like induc-
tive inferences than deductive ones (see Levinson
18 That q follows from p is simply introduced at step (3) without explanation.
One explanation is that in order to think that Pro’s assertion that p É q advances
Pro’s aim of stating p so q, i.e., advances Pro’s arguing for q, Pro should think
that q follows from p.
Inference Claims as Assertions 379
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
1983, pp. 114-116). However, p so q, but q doesn’t fol-
low from p, seems problematic when the argument
stated is an argument in its reason-giving sense. This
suggests that the follows-from claim is either a part of
what is said in uttering p so q used in a reason-giving
way or is a conventional implicature of this utterance
which isn’t cancelable (as described below). Either
way, this intuition counts against treating q follows
from p as a conversational implication of p É q, as-
suming that p É q is implicitly asserted in stating a
reason-giving argument p so q.
(b) Con’s reasoning in calculating the implicature of Pro’s
assertion of p É q is cogent even if Pro doesn’t believe
that q follows from p. This is because failing to fulfill a
maxim (or the Cooperative principle) may nevertheless
give rise to implicatures. Following Bach and Harnish
(1979, p. 167), one may quietly and unostentatiously
fail to fulfill a maxim, which is likely to mislead. In
such a case, Pro’s assertion that p É q is infelicitous,
precisely because it implicates that q follows from p
which Pro does not believe.
(c) Since an implicature is not part of what is said by the
utterance from which the implicature is inferred, it
may be neither true nor acceptable while what is said
is true or acceptable. E.g., p É q, may be true even
though q does not follow from p. Not satisfying Value,
distinguishes implicatures (both conventional and con-
versational) from implication.
Conversational implicature does not satisfy Belief. Suppose that
Pro states an argument, p so q, and thereby conversationally impli-
cates that q follows from p. Conversational implicature satisfies
Belief only if it is necessary that Pro believe that q follows from p
in order for Pro to implicate this. Given point (b) above, this isn’t
the case. Hence, conversational implicature fails Belief. Certainly,
Pro saying p É q as part of stating the argument while not believ-
ing that q follows from p is infelicitous since Pro is on the hook for
defending if not believing that q follows from p. However, this is
380 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
insufficient for conversational implicature satisfying Belief. As-
suming responsibility for believing the follows-from claim is short
of believing it. Supposing that Pro doesn’t believe that q follows
from p, it is nevertheless expressed as an implicature of Pro’s
utterance since Pro did not explicitly opt out of following the
maxims (or cooperative principle).
Given point (c), conversational implicature doesn’t satisfy Val-
ue. For example, what you utter may be true even though what you
implicate by your utterance is false. However, conversational
implicature satisfies Primary. Consider an utterance, I am on a
diet, made in response to the following question. Would you like
dessert? Plausibly, a primary aim of the speaker’s utterance is to
convey as a conversational implicature that the speaker does not
want dessert.
4.3 Conventional implicature
According to Grice (1989), the statement of an argument that
explicitly uses an illative expression (e.g., “since,” “therefore”)
conventionally implicates the corresponding inference claim.19
Bach defines conventional implicature as follows.
A proposition is a conventional implicature of an utterance just in
case (a) the speaker (speaking seriously) is committed to the truth
19 In an oft-cited passage, Grice introduces the notion of conventional implica-
tures in his well-known “Logic and Conversation” in order to distinguish them
from conversational implicatures which is his main concern:
In some cases the conventional meaning of the words used will de-
termine what is implicated, besides helping to determine what is said.
If I say (smugly), He is an Englishman; he is, therefore, brave, I have
certainly committed myself, by virtue of the meaning of my words, to
its being the case that his being brave is a consequence of (follows
from) his being an Englishmen. But while I have said that he is an
Englishman, and said that he is brave, I do not want to say that I have
said (in the favored sense) that it follows from his being an English-
man that he is brave, though I have certainly indicated, and so impli-
cated, that this is so. I do not want to say my utterance of this sen-
tence would be, strictly speaking, false should the consequence in
question fail to hold. So some implicatures are conventional, unlike
the one [a conversational implicature] which I introduced this discus-
sion of implicature. (1989, pp.25-26)
Inference Claims as Assertions 381
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
of the proposition, (b) which proposition that is depends upon the
(or a) conventional meaning of some particular linguistic device in
the utterance, but (c) the falsity of that proposition is compatible
with the truth of the utterance. (1999, p. 331)
Condition (b) grounds the fact that conventional implicatures are
not cancelable. Horn puts this as, “[they are] not CANCELABLE
without contradiction” (2004, p. 4). This accommodates the intui-
tion that there is something always wrong in uttering p so q, but q
doesn’t follow from p, when the argument is being used in a rea-
son-giving way. Also, an illative has to occur in what is uttered in
order for the corresponding inference claim to be conventionally
implicated. So, if statements of arguments convey inference claims
as conventional implicatures, then they must contain, explicitly or
otherwise, an illative expression.
The expressions “therefore” and “but” are often used to illus-
trate conventional implicatures (but see Bach 1999, discussed
below). Suppose (i) and (ii) are uttered in ordinary conversations.
(i) Beth is listening to music, but she is in the living
room.
(ii) Beth is listening to music; therefore, she is in the liv-
ing room.
The descriptive meaning of utterances (i) and (ii), what Grice
identifies with what is said, is the same: (iii) Beth is listening to
music and she is in the living room. If either conjunct in (iii) is
false, then so are (i) and (ii). Sample conventional implicatures of
utterances (i) and (ii) are: CIi – Beth is not usually listening to
music in the living room; CIii –Beth is in the living room follows
from Beth is listening to music. The CIs illustrate conventional
meanings of utterance (i) and (ii) in typical conversational settings.
Pro utters p, so q in a critical discussion. Treating the inference
claim, q follows from p, as an implicature of Pro’s utterance takes
it to be a meaning of the utterance that is not part of what Pro says.
Accordingly, in stating p, so q, Pro is not saying q follows from p.
If the inference claim is a conversational implicature of Pro’s
utterance, then it is a pragmatic meaning that arises from the inter-
action between the communicative goals of Pro and Con in their
382 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
critical discussion, and the conversational maxims guiding rational
communication. In contrast, if q follows from p is a conventional
implicature of Pro’s utterance, p so q, then it is a conventional
meaning of Pro’s utterance that stems entirely from the conven-
tional meaning of “so.” However, the meaning of “so” does not
contribute to what Pro says and so does not affect the truth or
falsity of what Pro says.20
Conventional implicature does not satisfy Belief, because con-
ventional implicature stems entirely from the conventional mean-
ings of lexical items or grammatical constructions occurring in the
sentence uttered. If my utterance of p so, q conventionally impli-
cates that q follows from p, then it does so regardless of whether or
not I believe my commitment to the inference claim. Therefore,
conventional implicature fails Belief.
According to Grice, the meaning of a conventional implicature
does not contribute to the truth-conditional content of the state-
ment that has the conventional implicature. Accordingly, conven-
tional implicature does not satisfy Value, since, like conversational
implicature, it does not necessarily preserve acceptability or truth.
For example, what is uttered may be true even though what is
conventionally implicated by the utterance is false. The oddity of
treating the value of the inference claim as irrelevant to the value
of the argument’s statement can be brought out by considering a
clearly invalid argument with true premisses and a true conclusion,
such as the argument: “Some politicians are men and some politi-
cians are corrupt, so some men are corrupt.” Would it be coherent
to say: “That’s true, but your conclusion doesn’t actually follow
from the reasons you give”? If the correct response is affirmative,
then Grice’s position is vindicated. If the correct response is nega-
tive as I think, then Grice’s position is refuted.
20 As is commonly noted (e.g., see Korta and Perry 2020), an internal tension in
Grice’s view of conventional implicatures results from Grice maintaining that
(i) a conventional implicature is a conventional meaning of an utterance, (ii) the
conventional meaning of an utterance contributes to the truth-conditional
content of what is said, and (iii) the meaning of a conventional implicature does
not contribute to the truth-conditional content of the corresponding utterance.
Claims (i)-(iii) are prima facie incompatible.
Inference Claims as Assertions 383
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
Moving to Primary, it is standard to think that the proposition
expressed by a conventional implicature is secondary to the main
assertion of a declarative sentence (e.g., see Abbott 2011, Potts
2007). Following Potts (2007, p. 667), at the discourse level we
find if one objects to an assertion of, say, (i) on the previous page,
one is not construed as having objected to CIi which is liable to
slip quietly to the common ground, unless the objector is explicit
that she objects to the CIi-content as well. This illustrates what
Potts calls the assertoric inertness of conventional implicatures
(2007, p. 672). That conversational implicatures are assertorically
inert is a reason to think that they do not satisfy Primary.
Before turning to assertion, it is worth considering Bach (1999,
pp. 338-343) who argues that conventional implicatures do not
exist. Obviously, if this is correct then this counts against constru-
ing an arguer’s inference claim as a conventional implicature of
the statement of the argument. Bach deploys what he calls the IQ-
test (the indirect quotation test) to make the case that conventional
implicatures are not distinguished from the descriptive meaning of
utterances and so are part of what is said in the sense of Grice. To
illustrate, between (ii) and (iii), which is an accurate paraphrase,
using indirect quotation, of what S uttered?
S: (i) Beth is listening to music on the stereo and therefore
Beth is in the living room
(ii) S said that Beth is listening to music on the stereo and
therefore Beth is in the living room.
(iii) S said that Beth said that Beth is listening to music on
the stereo and that Beth is in the living room.
According to Bach, it’s (ii). Statement (iii) leaves out the contribu-
tion of “therefore”; unlike (ii), it does not report S’s statement of
an argument. If (ii) is true, which seems right, then what S says is
(i). If S had just said that Beth is listening to music on the stereo
and that Beth is in the living room, then (ii) would be partly un-
true.
384 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
4.4 Assertion
As some have noted (e.g., Bach 1999, p. 330), Grice’s example of
“therefore” as a device of conventional implicature is unfortunate
since that q follows from p seems to be required for the truth of
what is stated by uttering p, therefore q. This suggest that, contra
Grice, q follows from p is part of what is said in stating p, there-
fore q. Hitchcock remarks that in stating an argument—in its
reason-giving sense—“[t]he arguer implicitly claims [italics are
his] that the conclusion… follows from the reason or reasons from
which it is drawn” (2011, pp. 191-192). Accordingly, by stating an
argument in its reasoning giving sense a speaker claims, i.e., as-
serts, the corresponding inference claim.
What distinguishes assertion from other types of speech acts?21
I favor an expressive account of assertion (for challenges see
MacFarlane 2011). Briefly, your utterance counts as an assertion
only if it expresses your belief of the propositions stated (Searle
1979 p. 12, Bach and Harnish 1979 p. 42). Expression of belief are
caused by the beliefs they express (Williams 2002, pp. 73-75,
Owens 2006). As noted by Williams (2002, p. 74), since only
sincere assertions can express beliefs in this direct way, insincere
assertions are assertions in a parasitic sense.22 Since you express
an inference claim by means of sincerely asserting it only if you
believe it, (sincere) assertion satisfies Belief.
Assertion satisfies Value. In arguing for a conclusion by stating
an argument, a speaker S performs the speech acts of asserting the
premises, the conclusion, and asserting the corresponding infer-
ence claim. Drawing on Bach (1999), Potts labels the phenomenon
of an individual sentence that is uttered expressing multiple mean-
21 For useful overviews of the many different accounts of assertion developed in
response, see Brown and Cappelen (2011b) and Pagin (2016).
22 According to Williams, “…the standard conditions of A’s asserting that P are
that A utters a sentence ‘S,’ where ‘S’ means that P, in doing which either he
expresses his belief that P, or he intends the person addressed to take it that he
believes that P” (2002, p.74). The second disjunct, which accommodates acts of
insincere assertion, expresses what I take to be the sincerity condition for A’s
act of conversationally implicating that P by uttering “S.” Again, as previously
suggested, A conversationally implicating that P does not require that A believe
P.
Inference Claims as Assertions 385
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
ings, semantic multidimensionality (2007, pp. 674-676). Constru-
ing the meaning of a sentence as the proposition it expresses,
semantic multidimensionality is instantiated by a one-sentence-
potentially-many-propositions semantics. This enables the analysis
of the meaning of a speaker S’s utterance, p, so q in terms of two
propositions expressed by the sentence S utters: p and q, q follows
from p. Here, what is expressed comprises more than one proposi-
tion. Since part of what is expressed in stating an argument is the
inference claim asserted, the inference claim must be true if what
is stated in the statement of the argument is true. Finally, assertion
accommodates Primary by the fact that the inference claim is
asserted, and what is asserted in an utterance is a primary point of
the utterance.
In sum, S’s assertion p, so q by means of uttering it asserts that
q follows from p, which is among the propositions expressed by
S’s utterance. Since S doesn’t utter q follows from p, the question
arises what determines whether the assertion of the inference claim
by means of uttering p, so q is implicit or explicit. Here I follow
Grice in understanding what is explicitly expressed by an utterance
strictly in terms of what is said by means of the utterance (see note
16, above). Defense of this substantive claim is beyond the scope
of this paper. Briefly, by stating an argument you explicitly assert
the corresponding inference claim only if you utter an illative
expression in stating the argument. Otherwise, the inference claim
is implicitly asserted.23
I’ve said that statements of arguments in their reason-giving
sense are truth-evaluable. However, as Bach notes (1999, p. 354),
a sentence that expresses more than one proposition, rather than a
conjunction of them, does not have a unitary truth condition. If my
utterance expresses several propositions all true except for one, is
what I say false? Without addressing this question directly, we can
say that what is uttered by stating an argument is true if every
proposition that is expressed is true.
23 I distinguish between implicit assertions and mere implications. Taking S to
be implicitly asserting an inference claim construed as a mere implication of her
statement of the argument threatens to erase what many take to be an intuitive
and useful distinction between what is asserted and what is merely implied (e.g.,
McFarlane 2011, p. 80).
386 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
That assertion satisfies Belief, Value, and Primary provides a
rationale for the paper’s central thesis: by stating an argument in
arguing for its conclusion in its core sense a speaker S asserts the
associated inference claim. Before concluding, I now sketch my
responses to two objections to this thesis.
One form of pushback questions the belief condition, which I
have labelled Belief. This criticism says that when one argues for a
conclusion one necessarily represents oneself as believing that the
conclusion follows from the premises given in its support. Of
course, merely representing oneself as believing p isn’t to believe
p. My initial response is two-fold. First, the criticism presupposes
that there is just one way of arguing for a conclusion, which I
reject. Again, my focus here is on arguing for a conclusion by
advancing an a persona argument. Surely, this is one way people,
in fact, argue. I take this to be arguing for a conclusion in its core
sense. Second, when one presents an a persona argument one
advances the premises as one’s reasons for which one believes the
conclusion. One doesn’t advance the premises so construed unless
one believes that the conclusion follows from them. I defend this
substantive claim in forthcoming work. A takeaway here is that
how we understand what counts as arguing for a conclusion mat-
ters to how we think the inference claim is conveyed by one’s
statement of the corresponding argument.
Another form of pushback undercuts motivation for the paper’s
central thesis by advancing either an alternative plausible story
about mere implication or a Neo-Gricean account of implicature
that satisfies Belief, Value, and Primary. For example, regarding
mere implication, one might follow Harman (1986) and argue that
one believes what is easily inferable from what one expresses.
Harman distinguishes between explicit and implicit belief. Briefly,
you believe something explicitly if your belief involves an explicit
mental representation whose content is the content of that belief
(p. 13). You believe something only implicitly if it is not explicitly
believed, but, for example, is easily inferable from one’s explicit
beliefs (p. 13). Given that you explicitly believe that Beth has at
most four children, one can easily infer that Beth does not have
five children, that Beth does not have six, and so on. So, all these
Inference Claims as Assertions 387
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
propositions are things that Beth believes implicitly.24 In response,
I say that one doesn’t argue for a conclusion by advancing one’s
reasons for it (i.e., doesn’t advance an a persona argument) unless
one explicitly believes the associated inference claim, i.e., unless
one possesses, in a belief-like way, a representation with that
content. Development of this response is beyond the scope of
this paper.
5. Conclusion
One way of arguing for a conclusion is by stating an argument for
it. How must the associated inference claim be conveyed? An-
swers depend on the operative sense of arguing for a conclusion.
Here I have focused on arguing for a conclusion in what I think is
its core sense. Arguing for a conclusion in this way involves ad-
vancing an a persona argument, i.e., an argument put forward as
the author’s own for the conclusion. I’ve motivated the conditions
Belief, Value and Primary by considering the connection between
arguing for a conclusion so construed and inference claims. Of the
alternatives considered here, I have argued that assertion best
satisfies Belief, Value and Primary as opposed to mere implication
and implicature. Accordingly, when stating an argument in order
to argue for its conclusion—in its core sense—one asserts the
corresponding inference claim as opposed to merely implying or
implicating it.
Acknowledgements
Thanks to the reviewer whose substantive and insightful com-
ments on an earlier draft greatly improved the paper. Also, thanks
to Michael O’Rourke for his comments on an earlier draft and
thanks to David Godden for discussions that helped my thinking
on several topics addressed in the paper. Of course, I am solely
responsible for any remaining errors.
24 For other ways that something can be merely implicitly believed see Harman
1986, pp. 13-14.
388 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
References
Abbott, Barbara. 2006. Where have some of the presuppositions gone?
In Drawing the boundaries of meaning: Neo-Gricean studies in
pragmatics and semantics in honor of Laurence R. Horn, eds. B.
Birner and G. Ward, 1-20. Philadelphia: John Benjamins.
Aikin, Scott and Robert Talisse. 2019. Why we argue (and how we
should). New York: Routledge.
Bach, Kent. 1994. Conversational impliciture. Mind and Language 9:
124-162.
Bach, Kent. 1999. The myth of conventional implicature. Linguistics and
Philosophy 22: 262-283.
Bach, Kent and Robert Harnish. 1979. Linguistic communication and
speech acts. Cambridge, MA: MIT Press.
Bermejo-Luque, Lilian. 2011a. Giving reasons: a linguistic-pragmatic
approach to argumentation theory. Dordrecht: Springer.
Bermejo-Luque, Lilian. 2011b. Exchanging reasons: Responses to
critics. Theoria 72: 329-334.
Biro, John and Harvey Siegel. 2006. In defense of the objective epistem-
ic approach to argumentation. Informal Logic 26: 91-101.
Brown, Jessica and Herman Cappelen. 2011a. Assertion: New philosoph-
ical essays. New York: Oxford UP.
Brown, Jessica and Herman Cappelen. 2011b. Assertion: An introduc-
tion and overview.” In Assertion: New philosophical essays, 1-17.
Copi, Irving and Carl Cohen. 2005. Introduction to logic, 12th ed. Upper
Saddle River, NJ: Pearson/Prentice Hall.
Corcoran, John (ed.). 1993. Editor’s introduction. In An introduction to
logic, 2nd, ed. Morris Cohen and Ernest Nagel, pp. xvii-xlvi. Indian-
apolis: Hackett Pub. Co.
Eemeren F. H. van and R. Grootendorst. 1984. Speech acts in argumen-
tative discussions: A theoretical model for the analysis of discussions
directed towards solving conflicts of opinion. Berlin/Dordrecht: De
Gruyter/Foris Publications.
Eemeren F. H. van and R. Grootendorst. 2004. A systematic theory of
argumentation: The pragma-dialectical approach. New York: Cam-
bridge UP.
Eemeren F. H. van and P. Houtlosser. 2003. The development of the
pragma-dialectical approach to argumentation. Argumentation 17:
387-403.
Epstein, Richard. 2002. Five ways of saying “therefore.” Belmont, CA:
Wadsworth/Thompson Learning.
Inference Claims as Assertions 389
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
Goldman, Alvin. 1994. Argumentation and social epistemology. Journal
of Philosophy 91: 27-49.
Govier, Trudy. 2010. A practical study of argument, 7th ed. Belmont,
CA: Wadsworth Cengage Learning.
Grennan, Walter. 1994. Are ‘gap-fillers’ missing premises? Informal
Logic 16: 185-196.
Grice, H. P. 1989. Studies in the way of words. Cambridge, MA: Har-
vard UP.
Groarke, Leo. 2002. Hilary Putnam on the end(s) of argument. Philo-
sophica 69: 41-60.
Hamblin, C.L. 1970. Fallacies. London: Methuen & Co Ltd.
Hitchcock, David. 2007. Informal logic and the concept of argument. In
Philosophy of Logic, ed Dale Jacquette, 101-129. Amsterdam: Else-
vier.
Hitchcock, David. 2011. Inference claims. Informal Logic 31: 191-228.
Horn, Larry. 2004. Implicature. In The handbook of pragmatics, eds L.R.
Horn and G. Ward, 3-28. Malden, MA: Blackwell Publishing Ltd.
Kelley, David. 2014. The art of reasoning: An introduction to logic and
critical thinking, 4th ed. New York: W.W. Norton & Company, Inc.
Korta, Kepa and John Perry. 2020. Pragmatics. In The Stanford encyclo-
pedia of philosophy (Spring Edition), ed. Edward N. Zalta. URL ac-
cessed 16 July 2020:
.
Levinson, Stephen.1983. Pragmatics. Cambridge: Cambridge UP.
MacFarlane, John. 2011. What is assertion? In Assertion: New philo-
sophical essays, 79-96.
Meiland, Jack. 1989. Argument as inquiry and argument as persuasion.
Argumentation 3: 185-196.
Neale, Stephen. 1992. Paul Grice and the philosophy of language. Lin-
guistics and Philosophy 15: 509-599.
Owens, David. 2006. Testimony and assertion. Philosophical Studies
130: 105-129.
Pagin, Peter. 2016. Assertion. In The Stanford encyclopedia of philoso-
phy (Winter Edition), ed. Edward N. Zalta. URL accessed 13 August
2020:
.
Pinto, Robert. 2001. The relation of argument to inference. In Argument,
inference, and dialectic, ed. R. Pinto, 32-45. Dordrecht: Kluwer Aca-
demic Pub.
Pinto, Robert. 2011. The account of warrants in Bermejo-Luque’s giving
reasons.” Theoria 72: 311-320.
390 McKeon
© Matthew McKeon. Informal Logic, Vol. 41, No. 3 (2021), pp. 359–390.
Potts, Christopher. 2007. Into the conventional-implicature dimension.
Philosophy Compass 2: 665-679.
Scriven, Michael. 1976. Reasoning. New York: McGraw-Hill Book Co.
Searle, John. 1969. Speech acts: An essay in the philosophy of language.
Cambridge: Cambridge UP.
Searle, John. 1979. Expression and meaning: Studies in the theory of
speech acts. Cambridge: Cambridge UP.
Sinnott-Armstrong, Walter and Robert Fogelin. 2010. Understanding
arguments. Belmont, CA: Wadsworth Cengage Learning.
Thomson, Judith Jarvis. 1967. Reasons and reasoning. In Philosophy in
America, ed Max Black, 282 – 303. Ithaca, NY: Cornell UP.
Vorobej, Mark. 2006. A theory of argument. Cambridge: Cambridge UP.
Williams, Bernard. 2002. Truth and truthfulness. Princeton: Princeton
UP.