Goodwin paper


Argument Has No Function     69

Argument Has No Function

Iowa State UniversityJEAN GOODWIN

Abstract: Résumé: Douglas Walton a eu raison
d’attirer notre attention sur les aspects
pragmatiques des arguments. Toutefois,
il a insisté qu’on devrait interpréter et
évaluer des arguments en prenant compte
de leurs fonctions. Je suis en désaccord
avec ceci: on ne peut pas déterminer les
fonctions des arguments dans le sens
dont Walton a besoin, et même si on le
pouvait, on ne pourrait pas fonder les
normes de l’argumentation sur ces
fonctions. Je propose comme alternative
à la théorie fonctionnelle de la
pragmatique argumentative une théorie
qui concentre sur la façon que des
participants planifient stratégiquement et
imposent des normes sur eux-mêmes afin
de renforcer leurs arguments.

© Informal Logic Vol. 27, No. 1 (2007): pp. 69-90.

Keywords:  argument; argumentation; argument-evaluation; informal-logic; Walton,
Douglas; functionalism; dialogue; dialectics; pragmatics.Argument has no function

1. Introduction

                 Douglas Walton has been right in
calling us to attend to the pragmatics of
argument. He has, however, also insisted that
arguments should be understood and
assessed by considering the functions they
perform; and from this, I dissent. Argument
has no determinable function in the sense
Walton needs, and even if it did, that function
would not ground norms for argumentative
practice. As an alternative to a functional
theory of argumentative pragmatics, I
propose a design view, which draws
attention to the way participants
strategically undertake and impose norms
on themselves in order for their arguments
to have force.

Over the last twenty years, Douglas Walton has been leading the way to the pragmatic
study of argument. Throughout his extensive body of work he has urged us to
look at actual arguments, insisting that the goodness of an argument can only be
assessed in its real-world context. He has called on us to recognize: that an arguer
makes an argument for a purpose—not uncommonly, because she is obliged to
address an issue between herself and others; that her obligations at any moment
depend in key ways on what she and others have said before; that as argumentative
talk proceeds, her argument often ramifies beyond a simple premise/conclusion
pairing into more complex patterns; that the situatedness and complexity of
argument legitimately leaves room for the arguer to make strategic choices in the
design of what she does. Walton has also reminded us that our job as theorists is to
use care in sorting out and systematizing the patterns in argumentative talk (strategies,
schemes, speech acts etc.), and good judgment in assessing its quality.



70     Jean Goodwin

This pragmatic approach has opened new questions for theory. What kind of
story should we tell about the relationship of an argument to its context? How
should we conceptualize or model “context”?—what are its most relevant features?
What are the norms by which arguments-in-context should be evaluated?

Consider this as an example—one I’ll be using as a test case occasionally in
this paper. It is widely recognized by both theorists and practitioners of argument
that in some contexts, some arguers have a burden of proof: an obligation to
produce arguments of a certain quality. For instance, at times an arguer is obliged
to reply to objections of some sort. In such a case, to be good, an argument must
(among other things) meet the contextually-given obligations. So we need to ask:
Where do probative obligations come from? What features of the context give rise
to them? It seems obvious that prior talk is often relevant to establishing burdens
of proof, while the phase of the moon is not. But why? And what aspects of the
prior talk? Context is important. But a pragmatic theory of argument must go on to
give an account of how, in at least some specific contexts, the burden of proof is
in fact assigned.

Walton has led the way in this, as well. He has made what I am going to call a
function claim, proposing that the most important feature of the context of an
argument is the social function that the surrounding talk is performing. In this
paper, I contend that function claims lead us astray. Argument is indeed useful. But
argument has not been shown to have any function, in the specific sense I am
using it here; and even if it does have a function, we cannot derive norms from
that fact. We need to stop making function claims, and take up other ways of
conceptualizing context. I close by proposing one such alternative.

2. Function claims: what, who, why

All would probably agree that argument is functional, in a broad sense. People
aren’t insane when they make arguments. In general, they are trying to get something
done, and they expect their arguments make a difference. All discourse is functional
in this sense (e.g., Sanders, 2005), and argumentative discourse is no exception.
Many of us, furthermore, think good argument is necessary for the human species
to flourish, or even survive. And so we might say that argument performs a vital
function.

I strongly endorse the functionality (purposiveness, usefulness, value,
effectiveness, dignity) of argument in this broad sense. The functionalism that I
am objecting to, by contrast, is a specific mode of theory construction—a specific
way of modeling the general functionality of argument. For the purposes of this
paper, an explicit function claim consists of the following complex of assertions:

(a) The context of an argument should be conceived as a joint activity.1

(b) That joint activity has the function of achieving a social good.
(c) The norms of argument include those rules (principles, values,



Argument Has No Function     71

standards, etc.) an argument must follow (live up to, instantiate, meet,
etc.), in order for the joint activity in which it is embedded to achieve its
function.

Douglas Walton has been one of the most prominent theorists making a function
claim, as the following brief review of his theory reminds us.

(a) Arguments, for Walton, occur in the “standard contexts” (1995, 98) provided
by “joint activities” (1998, 29) called dialogues. As he defines it,

a dialogue is a conventionalized framework of goal-directed activity in which
two participants interact by taking turns to perform speech acts. . . . The
various speech acts are linked together in a sequence that has a purpose and
direction as the dialogue proceeds. The purpose is determined by the goal of
the dialogue as a recognized type of social activity (1995, 98; see also, e.g.,
2006, 2; 1998, 29).

A dialogue thus provides a mutually recognized, shared framework or structure
within which the individual activities of the arguers take place. As is well known,
Walton envisions multiple types of dialogue; a standard list includes persuasion
dialogues, negotiations, deliberations, inquiries, information-seeking dialogues and
quarrels (e.g., 1998), but in other places he mentions up to eleven types (1992,
95), with further divisions into subtypes possible.

(b) Participants in a dialogue have their own individual aims, but these, Walton
insists, must be distinguished from the collective goal that defines the dialogue
itself (Walton & Krabbe, 1995, 67; Walton, 1995, 101). “What is important” for
argumentation theory, he says, “is not the particular arguer’s purpose but the
purpose of the dialogue as a whole” (1995, 271). The collective purpose. goal or
aim of a given type of dialogue is to realize a recognizably valuable state of affairs
shared by the participants—that is, a social good. A persuasion dialogue aims at
resolving or clarifying issues; an inquiry aims to prove or disprove a hypothesis; a
negotiation aims at a reasonable settlement that both parties can live with; deliberation
aims to decide the best available course of action to be taken (2006, 183). Even a
quarrel achieves a social good. Individual participants may want to hurt each other,
but the dialogue can “reveal [the] deeper basis of conflict” (2006, 183), in the hope
that once “hidden grievances” are “exposed explicitly, acknowledged and dealt
with,” the “continuance of a personal relationship” will prove possible. “During a
good quarrel, the participants ‘make up,’ . . . vowing to be more thoughtful or
sensitive about this particular issue in future conduct (1995, 109). The quarrel, in
Walton’s view, thus performs a “cathartic function” (2006, 190).

(c) A dialogue provides a normative framework for arguments, and Walton
derives the norms from the collective goal of the dialogue. Indeed, Walton argues
that it the normative force provided by a dialogue’s goal-directedness that
differentiates his approach from the formal dialectics of Hamblin and others. Such
dialogue games, although interesting, fail to model well real-world argumentative
discourse because they are not “motivated” by the purposes actual dialogues are



72     Jean Goodwin

aimed to achieve. It is the collective goal of the exchange, Walton explains, that
allows us to pin down the exact dialectical norms that bear within it, among all the
possibilities formal systems offer (1995, 8).

How does a dialogue’s collective goal determine the norms that bear within it?
Most generally, norms for argumentative talk arise because participants in a dialogue
ought to cooperate to achieve the collective goal. In various works, Walton applies
this general principle to a variety of specific questions within argumentation theory.
For example, the dialectical rules of procedure are just those that “guarantee or at
least facilitate the reaching of [a given dialogue’s] goal” (Walton & Krabbe, 1995,
66). Argument assessment depends on the collective goal: to say that an argument
is good “in the functional or pragmatic sense . . . means that the argument is good
or useful in contributing to the goal of the dialogue” (1998, 43). Fallacies are
given a “functional” account; a fallacy is committed when an argument scheme is
used “in a manner that fouls up” a dialogue, “block[ing] or subvert[ing]” its goal
(1995, 22-23). An argument has a linked as opposed to convergent structure if all
its premises must “function together” in order to achieve the argument’s purpose
in the context of dialogue (1996a, 176-77). Relevance is established when an
argument is locally or globally appropriate to achieving a dialogue’s goal; irrelevance,
when an argument is “unproductive” for it (2006, 272). Or consider finally Walton’s
derivation of the burden of proof. He comments:

The concept of an obligation applies to all types of dialogue. The obligation
is the function the participant has to perform, according to the rules of the
dialogue, in order to fulfill his goal in the dialogue. Burden of proof is a
subcategory of obligation. . . .  Burden of proof is an important and useful
idea where conclusive resolution of a disputed issue by appeal to decisive
evidence (knowledge) is not practical or possible. The problem, in such a
case, is that argumentation could go on and on and never reach a resolution.
Burden of proof is a practical solution to this problem which works by setting
a required weight of strength of argumentation as sufficient to prove
(disprove) the contention, and thereby close the dialogue off from further
argumentation (1996b, 24).

In sum, Walton puts forward what he himself calls a “functional view of
argument” (1992, xii), (a) locating arguments within the contexts of dialogues of
specific types, (b) dialogues which are aimed to achieve specified social goods,
(c) a social goods which ground context-dependent norms for arguments.

Walton is not alone in making a function claim. He is joined most notably by the
pragma-dialecticians. As is well known, they propose (a) that an argument is to be
“view[ed] . . . as part of a critical discussion.” A critical discussion (b) is a speech
activity “aimed at resolving a difference of opinion.” It is regulated (c) by a “valid
dialectical procedure,” where validity

depend[s] on how efficient and efficacious it is in furthering the resolution of
disagreement. . . . A valid dialectical procedure makes it possible to determine
whether the argumentative discourse as it is conducted can be instrumental
in resolving a difference of opinion.



Argument Has No Function     73

So as a sort of shorthand they say: “argumentation has the general function of
managing the resolution of disagreement” (van Eemeren et al., 1996, pp. 278, 280,
278, 277).2

And Ralph Johnson is another. “To understand the product that we call an
argument,” he says, (a) “it is necessary to situate it within its proper context: the
practice of argumentation.” Unlike a dialogue or critical discussion, the practice of
argumentation is not a specific event (or process) of argument exchange; instead,
in the terms I am using here, the practice of argumentation is a joint activity at the
most general level. It is, in Johnson’s words, the widely recognized, socially
instituted “sociocultural activity” of exchanging arguments, the “shared
understanding” of which makes specific events intelligible. The practice of
argumentation, furthermore, (b) achieves certain goals “internal” to it, including:
“an increase in rationality and specifically a deeper understanding, and-or being
rationally persuaded, and-or coming closer to an acceptable position.” Finally, (c)
“a good argument is one that fulfills its purpose” by meeting the requirements
Johnson establishes as necessary to achieve these internal goods (2000, 154, 155,
181).

Other theorists, while not articulating full function claims, do use language that
echoes one or more of assertions (a), (b) and (c). This suggests that they may be
committed to functionalism, or at least are tending in that direction. Signs that may
indicate an implicit function claim include:

(a)-like talk: Naming the joint activity in which arguments are said to occur.
Many theorists start theorizing the context of argument by conceiving the context
“whole,” taking a top-down approach in line with (a). For these theorists, the most
interesting thing about argument is that it occurs within the context of a __________
(insert noun here). For example, Freeman contends that the “basic dialectical
situation can serve as a model of argument” (1991, 20). Such a baptizing of a
context suggests that a theorist may be leaning towards a function claim.

(b)-like talk: Using purpose-language without a purpose-er. Some theorists
ascribe aims, goals, ends, purposes (etc.) not to the people using arguments, but
to the joint activity or argument itself. For example, Michael Gilbert, who is generally
quite precise in referring to arguers’ goals, occasionally lapses into such language,
asserting that “the aim of coalescent argumentation is to bring about an agreement
between two arguers based on the conjoining of their positions in as many ways as
possible” (1997, 70). Similarly, Trudy Govier remarks that “the purpose of an
argumentative exchange is to communicate information, beliefs, and opinions [etc.]”
(1987, 149). As Jon Elster (1982) has remarked, “functionalist thinkers
characteristically use the passive voice;” when an argumentative activity is said to
aim at something, a function claim may be lurking.

(c)-like talk: Invoking an obligation to cooperate. An asserted function of
argument can be taken to regulate the behavior of individual arguers by way of
their alleged obligation to cooperate in a joint activity to achieve it; cooperation is



74     Jean Goodwin

the glue binding an empirical function to a norm of argumentation. Walton frequently
invokes the name of Grice in this regard, quoting “Cooperative Principle” (e.g.,
1998, 5; 1996a, 23; 1995, 255; 1992, 84), and other theorists have done the same
(e.g., Govier, 1987, 156).3 Invocations of an obligation to cooperation should
count as signs of a possible function claim .

Given their apparent popularity, we have to grant that function claims have
their attractions. First, a functional approach solves many problems in theory,
quickly. Instead of building a theory of context bit by bit, figuring out which
aspects of context are important for which aspects of argument and why, a function
claim conceives of the context of an argument entire, as a gestalt or organism;
only later need it work from top down to deal with the messy details. Second,
function claims resonate well with the apologetics we use to defend the legitimacy
of our enterprise. In order to persuade (say) students and administrators of the
importance of the argumentation course, we like to proclaim how useful—indeed
vital—it is for achieving all sorts of important objectives. (And, as I said above,
we’re right.) Finally, it’s just easy to talk function talk. It is a vocabulary that
seems ready to hand when we want to discuss social life.

In fact, the vocabulary is ready to hand; we are far from the first to adopt it.
The earliest attempts to explain social activity in terms that would meet the standards
of modern scholarship adopted a functionalist perspective. By the 1950s and 1960s,
however, early functionalism in sociology and anthropology had come under intense
critique, and by the late 1960s was at least temporarily defunct.4 In the following,
I draw from many of the issues that were raised in this debate to argue that we too
should abandon naïve function claims.

There are problems with each of the three component assertions of a function
claim. Start with (a). Personally, I have what might be called ontological doubts
about the alleged joint activities within which arguments are said to be used.
Arguments are everyday things; we’re intimate with them. So it seems odd to me
to explain them by reference to abstract social institutions, some of which do not
even have names in ordinary language (see also Dore’s (1967) critique of the
tendency to treat “institutions . . . as ultimate units”). Further, the move in (a)
seems suspiciously like assuming the conclusion: instead of working his way to
more and more sophisticated theories of context, the theorist finds that everything
he needs to know about the context of argument is already implicit in the conception
of the joint activity. This is theory without tears.

It is beyond the reach of this paper, however, to question the conditions for
saying that an institution—what Durkheim called a “social fact”—really exists.
Therefore I am going to focus my objections here on moves (b) and (c) of the
function claim. In the following, I argue that there has been no showing that the
(assumed) joint activities within which arguments occur have the asserted functions
(anti-(b)). And even if there are such functions, they are not sufficient to justify
specific norms of argument (anti-(c)). But that’s not a problem, I will conclude,



Argument Has No Function     75

since there is at hand another way of conceptualizing context—one which does
not make a function claim.

3. Argument has yet to be shown to have a function

2.1 Where is the evidence?
In move (b), theorists like Walton making function claims assert that the joint
activities within which arguments are embedded perform specified functions. To
say that something performs a function is to say it normally or reliably does
something, that it has consequences, in particular, socially useful consequences.
To borrow a more precise definition from Robert Merton, one of the most
sophisticated of the functional sociologists: “Functions are those observed
consequences [of a social institution] which make for the adaptation or adjustment
of a given [social] system” (1957, 51). Thus for Walton, argumentative dialogues
“adjust” our social system by promoting rational consensus, mutual understanding,
sound decisions and solid relationships. Other theorists noted above postulate
something like “rational dispute resolution” to be the (or a) function of argument;
so they appear to think that a reason-based consensus is a regular consequence of
some argumentative activities.

So my question is: How do these theorists know that the joint activity in fact
produces socially useful consequences like this? Where’s their evidence? Shouldn’t
it be easy, or at least possible, to show a significant case in which an argumentative
joint activity has promoted mutual understanding, rational consensus and so on?
Or alternatively, some statistics suggesting a correlation between an increase in
argumentative talk in some society, and an increase in mutual understanding (etc.)
there?

Unfortunately, no such evidence is being produced. Quite the contrary: move
(b) typically occurs at the very beginning of theorizing; theorists appear to treat it
as so obvious as to need no defense (either that, or it is a matter of stipulation—a
possibility I will return to in 2.4.3 below).

Early functional analysis in sociology sank in part because of just such
complaints. Even Merton himself decried the fact that sociological functionalists
“do not typically carry through operationally intelligible procedures, [and] do not
systematically assembled needed types of data” (1957, 49). Others complained
that functions were established “‘intuitively by the structure of the observer’s
language, or are assumed to be in nature’” (Davis, 1967, 387, quoting Buckley,
“Structural-Functional Analysis in Modern Sociology”).

The lack of evidence might be forgiven. At least some of the claimed functions
of argument are plausible, and if we had no reason to believe that they were
wrong, we might continue to accept them as the best currently available accounts
of what argumentative joint activities do. But we do have evidence of other
possibilities, or at least can imagine some. Consider three sources of evidence:
sociological studies, ordinary experience, and speculation.



76     Jean Goodwin

2.2 Some other consequences of argument
(1) Argument is functional, but in odd ways. There are a handful of empirical
studies examining the function of argumentative activities, suggesting several
possibilities. Deborah Schiffrin’s early “Jewish Argument as Sociability” (1984),
for example, gave evidence that argument can be an activity participants engage in
for its own sake and to maintain their relationships, not to resolve substantive
issues. Gary Fine’s (2001) study of high school debaters supported the idea that
their argumentative activity allows bright adolescents to experiment with the line
between acceptable and unacceptable behavior, for example by trying out arguments
which are outrageous but (within the local rules) valid.

Unfortunately, neither of these documented functions are among those thus far
proposed by argumentation theorists, nor do they seem useful for producing the
kinds of norms theorists want to produce in move (c) of the function claim.

(2) Argument is dysfunctional. So turn to ordinary experience. The most salient
consequence of the joint activities involving arguments is to make the participants
mad: this is at least a widely held ordinary view of the function of argument. It is
easy to elicit this view from students (Goodwin, 2005b)—as I am sure most of
my readers would agree. Indeed, a connection between “argument” and “war”
may be built into our very language, with its conventionalized metaphors like
“attacking” and “defending” positions (thus Lakoff & Johnson,1980; see also the
covers of several well-known argument texts). And even a linguist like Deborah
Tannen proceeds as if the “argument” aspects of the “argument culture” lead
inevitably to confrontations and fighting (1998).

But if increased hostility is the most salient outcome of the joint activities in
which arguments arise, then these activities are not so much functional as
dysfunctional, decreasing the “adaptation or adjustment” of the social system.
Hostility is not a social good but a social ill; we do not want to promote hostility by
establishing norms.

(3) Argument is consequential, but not functional. One of the leading alternatives
to functionalist sociology in the 1960s was a Marxism-inspired “conflict theory.”
In this view, social institutions are not in place because they are functional; they
are there (to speak very generally) because those with power have put them there,
to help maintain their power.

So consider this speculation—a story about argumentative activities that runs
along the parallel lines. In our society, the resources for making arguments are not
equally distributed. The rich and powerful have more access to the education,
information and self-confidence necessary to argue well. So, we can imagine,
they institute the exchange of arguments as a means of resolving disputes, since
these exchanges are games they are likely to win.

The powerful also have an interest in maintaining an ideology or false
consciousness that “dialogues reach rational resolution” and that “argumentative
activities benefit all.” That way, when the powerless lose the argument—as, being



Argument Has No Function     77

handicapped, they often will—they will blame themselves, not the social injustices,
for their defeat.

We can even project how argumentative “norms” could arise on this view.
Return for example the burden of proof as articulated within forensic debates in
U.S. colleges: anyone who proposes a change in the status quo is obliged to make
a prima facie case for that change. As Tom Goodnight pointed out at the inaugural
Alta conference, this burden of proof assignment instantiates a specifically
conservative political philosophy (1980). One could hypothesize that it is imposed
by those currently victorious in the social struggle as a repressive measure against
those who are currently losing in it. After all, the weight of the obligation explicitly
rests on those who would change things.

Lest this account seem too far-fetched, let me point out that similar objections
have been put forward, especially by feminist scholars, in opposition to current
political theories that stress deliberation (e.g., Sanders, 1997; Young, 1996). In an
unjust society, what purports to be a cooperative exchange of reasons really
perpetuates patterns of oppression.

Like the functional account, this hypothetical “conflict theory” conception of
joint argumentative activity stresses its social consequences: here, the consequence
of maintaining the powerful in power. This consequence is neither functional nor
dysfunctional, however; it does not either work to the benefit or detriment of
society as a whole.  That is because a “conflict theory” does not conceive of
society as being a whole, as naïve functionalism does. Instead, argumentative
activities are seen as benefiting some groups, hurting others. From this point of
view, function claims are not just false, they are ideologically driven, inherently
conservative attempts to mask injustice—a common complaint against functionalist
theories (see Demerath & Peterson, 1967 and Abrahamson, 1978).

And so again, we have a plausible (if speculative) account of what joint activities
involving argument can do, but one that doesn’t assert that they achieve a social
good. Given the existence of these three alternatives—the odd functions,
dysfunctions and non-functions of argumentative activities—it seems reasonable
to ask those making function claims to produce some evidence in support of their
alleged function(s).

2.3 The special problems of multiple joint activities
Before turning to answer some objections, let me insert an aside about Walton’s
pluralistic version of the function claim.

To some extent, this approach seems to avoid my complaints about lack of
evidence, since whatever function I can document, experience or imagine, a multi-
functionalist can probably absorb. Yes, Walton may say, arguments can contribute
to joint activities which resolve disputes; but they can also contribute to emotional
release—and, as far as we know, to sociability and class warfare as well.



78     Jean Goodwin

My question to multi-functionalists like Walton is instead: given this wide range
of possible joint activities or “dialogue types,” how do you know which one is
being actualized in any specific argumentative transaction? For unless the collective
goal of a joint activity can be securely identified, the norms governing it cannot be
derived and applied.

Walton is in fact quite candid about this problem. The dialogue type in play at a
given moment “may not be clearly specified or apparent” (1992, 43). Further, a
dialogue may be of one type, but have a “flavor or overtone” of another (Walton &
Krabbe, 1995, 70); it may even be a mixture of several types (e.g., 1998, c. 9).
Indeed, some of our most prominent actual argumentative joint activities—
institutions which are named in ordinary language, possess explicit rules, and sustain
rich normative practices—institutions which even an institution-agnostic like myself
finds hard to doubt—these very institutions Walton finds to be the most mixed,
according to his typology. “Political debate” (that is, formal debate in a legislature)
“is typically a complex mixture of all six types of dialogue,” Walton says (1995,
128). Similarly, the closing arguments of trials in the common law system, while
being mostly persuasion dialogues, also have elements of expert consultation
dialogue, negotiation, and probably eristic as well. But if norms are to be derived
from the collective goal of a dialogue, how are we to determine what they are
when so many goals are in play? Although of course “special sensitivity and
qualifications” will always be required to discern and apply norms of argument to
complex, real-world situations (1998, 218), as functionalism becomes too multi-,
it loses its explanatory power.5

It’s worth noting that this is not just a problem for argumentative theorists.
Arguers in practice also need to figure out what norms they should be following.
As Walton confesses, given the multiple possible goals of argumentative joint
activities, “participants in argumentation are often unclear as to what type of dialogue
they are supposed to be engaged in” (1995, 129), or if clear, they may disagree
(1995, 123). So how do participants come to a mutual understanding about which
dialogue type they are pursuing? (As in the Monty Python sketch, what if one
person wants an argument, but the other thinks he needs abuse?) Walton occasionally
speaks as if participants in an argumentative joint activity start by reaching an
agreement about what kind of dialogue they will have (e.g., 1998, 205). Since
argumentative talk often begins because people disagree, it seems unrealistic to
assume that they will reach such procedural agreements easily. For example, in
many contemporary controversies each side claims that the other has the burden
of proof (Gaskins, 1992). Who is right? Presumably, participants in an argumentative
dialogue will need to have a meta-dialogue to establish the goal and thus the norms
of their joint activity. But what type of dialogue will that be? An infinite regress
seems to threaten.

In fact, we solve this sort of problem in practice all the time. I call a student
into my office, or I get called in by my dean. What kind of conversation are we



Argument Has No Function     79

two going to have? In general, in our society the higher status person can try to set
the tone—”waiving” formality, for example, or starting angrily. The other may go
along—or may not; one can refuse to sit when invited, for example. The norms
and other expectations governing a particular interaction arise out of the strategic
actions of the participants themselves.

Examples like these suggest the following thought. Even if we assume that
arguers have a set of joint activities and associated norms “at hand,” each will still
need strategies for bringing his desired activity-cum-norms to bear on the immediate
situation. These strategies are not part of the joint activity; they need to be theorized
on their own (for some indications of how to do this, see Section 4 below). So a
functionalist theory, even in its multi-functionalist variation, remains at a minimum
radically incomplete.

2.4 Some objections answered
(1) It could be said that functions like “sociability” and “enabling experimentation”
are merely parasitic on the “real,” “central,” or “paradigmatic” function of
argumentative talk, that is, rational resolution of disputes (or whatever). So the
existence of these odd functions does not undermine move (b) of the function
claim.

Fine! But note, that should make it even easier to find some evidence that
rational resolution (or whatever) is the, or even a, regular consequence of
argumentative talk, given that it is so central and paradigmatic.

(2) It could be said that the ordinary views and my “conflict” hypothesis do
not undermine the existence of a function for argumentative talk. For following
functionalist theorizing in sociology (e.g., Merton 1957), the “manifest” and “latent”
functions of the joint activity could be distinguished. The “manifest” functions are
those apparent to participants, as for example the function of making people mad;
hidden from them, although visible to the skilled investigator, are the real, “latent”
functions the activity achieves. It could even be that the hidden, latent functions of
an activity are achieved precisely because of its manifest dysfunctionality, by a
sort of “invisible hand” (more on this below). Indeed, I’ve heard the pragma-
dialecticians say just this: the more hostile arguers are—the more they want to
defeat each other—the more they are motivated to produce high quality arguments,
and thus the more functional their joint activity becomes. So the existence of
dysfunctions or conflicts does not undermine move (b) of the function claim.

Again, fine! But to assert that something is the latent function of an activity
suggests that it is even more important for the skilled investigator to produce some
evidence for its existence, given that the function is by definition not apparent to
most of us.

(3) Finally, it could be said that argumentation theory is not a branch of sociology;
in making function claims, we are not trying to describe or explain a social activity,



80     Jean Goodwin

we are trying to develop norms for it. Whatever the observed consequences of the
joint activities within which arguments are embedded—whether there are any
observed consequences at all—still, arguers ought to pursue the activities in order
to achieve the social goods the theorist posits. As Walton puts it, “the structures of
dialogue” he proposes

are normative models that represent ideals of how one ought to participate in
a certain type of conversation if one is being reasonable and cooperative. It
is important not to confuse this with an account of how participants in
argumentation really behave in instances of real dialogue that take place
(Walton & Krabbe, 1995, 67).

Similarly, the pragma-dialecticians recognize that the function of argument is
an idealization: a tool for critiquing argument activities, weeding out the dysfunctional
and nonfunctional growths, and fostering whatever among the actual activities is
worth salvaging. And Govier likewise has warned us that to say “that the practise
or social institution of argument has [rational persuasion] as its typical function or
purpose” is not to make a “straightforwardly empirical” point, but a “conceptual
and normative one” (1987, 150).

But an idealization or a conceptual point about actual argumentative discourse
still needs defending. For example, if rational resolution is indeed the end of some
joint activity involving argument exchange, shouldn’t it be paradoxical in the course
of such an activity to say something like “I am making an argument (that p), but I
am not trying rationally to resolve the dispute with you (that p)”? To me, however,
that assertion does not sound strange. Indeed, although I am here making arguments
that argument has no function, I am not trying to resolve the dispute about
argument’s function. That would be a hard ambition, since I do not expect to
change the mind of anyone who has committed himself to a function claim. (I do,
however, expect to force theorists to take responsibility for what they say, and
thus to give them strong encouragement to think. And, being an optimist, I hope
that thinking will lead them to the correct answer.)

Or if not by such a demonstration of pragmatic incompatibility, a theorist might
defend an idealization by displaying its power in understanding and assessing
argumentative talk. Along these lines, he might demonstrate how useful it is to
treat some actual argumentative activity as if it were a “persuasion dialogue” or
“critical discussion”—how adopting the model enables him to make a more
sophisticated critique of the activity than a reasonably skilled, but theoretically
naïve, practitioner could. Or again, a theorist might defend an idealization by pointing
to evidence that arguers in practice are orienting themselves towards the asserted
norms, accumulating empirical evidence that would suggest their force.

Thus far, however, theorists making function claims have yet to muster such
defenses of their idealizations. Of course, an idealization could still be interesting;
it could allow theorists to explore systematically and precisely the consequences
of making certain simplifying assumptions. It is possible to proceed to construct



Argument Has No Function     81

theory by stipulation, saying: “Let us call units of discourse which contribute in
such & such a way to achieving such & such a collective goal, ‘arguments.’”  But
if that’s what they are trying to do, theorists making function claims run the risk of
misleading their readers into thinking that they are discussing, well, arguments—
the familiar, everyday kind (see Godden, 2005 for a similar point).

But Walton in particular doesn’t want the joint activities he posits to be just
idealizations, like formal dialogue games. Walton finds it attractive to model
argumentative activities as dialogues specifically because it seems to allow
descriptive/empirical and normative/conceptual approaches to be linked (e.g.,
Walton, 1998, 29; Walton & Krabbe, 1995, 175). And I believe that other theorists
making function claims follow him in this. If so, it seems incumbent on them to
show that the goal-directed joint activities they posit indeed articulate well the
ideals implicit in actual argumentative talk.

Finally, let us assume that this task can be accomplished: that the idealizations
can receive a partial defense. Still, I will argue, even as an idealization a collective
goal won’t do the job function theorists expect of it: it won’t ground norms
governing argumentative activity. To fill out that idea, I move from function claim
(b) to claim (c), and open a new section.

3. Even if argument has a function, we cannot derive norms from that
    fact

3.1 “Ought” from “is”?
In move (c), theorists making function claims derive specific norms, binding on
arguers, which purport to be necessary for the joint activity constituting the context
of arguments to achieve its assigned collective goal. In principle, however, there
are problems with making this move.

If the joint activity which provides the context for argument achieves, really or
ideally, a social good, then certainly if we happen to be designers of social institutions
we will (other things being equal) want to create conditions within which that
activity will flourish. That is straightforwardly a matter of prudence.

What is not straightforward is how this prudential concern in the design of
institutions translates into norms binding on individual participants in those
institutions.6 This seems especially true if the alleged function of the argument
activity is latent, that is, unrecognized by the participants. Every arguer undoubtedly
has his own purposes. But why should he be obligated to so act as to achieve the
purpose of the dialogue, especially if he is not even aware of that purpose? As an
analogy: According to the “invisible hand” theory of economic activity, individual
greedy behavior performs the important social function of creating wealth. If so,
those setting economic policy are wise to take this fact into account, harnessing
greed for the common good. Nevertheless, individual economic actors are not
obliged to be greedy. Part of the excitement of this mode of theorizing is the way



82     Jean Goodwin

that it transforms individual vice into collective good. But vice remains vice; a
functional account of an activity does not seem to impose norms on individual
actors. You cannot have at once a functional theory, especially one that relies on
the “invisible hand,” and one that derives norms.

3.2 The possibility of functional alternatives means that functions do not determine
      norms

Let us assume, however, that those making function claims can overcome this
difficulty. After all, it does seem plausible that individuals are in some fashion
obliged to promote the common good, which suggests that at least manifest functions
have some normative punch. But there is still a second difficulty facing move (c):
showing that this specific set of norms is necessary to achieve the function.

Early functionalism in sociology and anthropology attempted to insist on the
functional indispensability of certain social arrangements—asserting, for example,
that every society needed to have a religion of some sort to establish common
values. This got the theorists into the obvious trouble of explaining how societies
had such different religions, and some no apparent religion at all. More sophisticated
functionalism, by contrast, recognized that while the social function might be
indispensable, the social institution was not. In Merton’s summary:

The same function [may] be diversely fulfilled by alternative items.
Functional needs are here taken to be permissive, rather than determinant, of
specific social structures. Or, in other words, there is a range of variation in
the structures which fulfill the function in question. In contrast to [the] concept
of indispensable cultural forms (institutions, standard practices, belief-
systems, etc.) there is, then, the concept of functional alternatives (1957, 33-
34).

What does this mean for the joint activities which are the context for arguments?
While it may be that society needs to promote (say) mutual understanding somehow,
we can expect there to be many functional alternatives for accomplishing this. No
single set of norms, therefore, can lay claim to being necessary for achieving the
desired function. Thus the move in function claim assertion (c), from general
function to specific norms, is illegitimate.

Consider this analogy. To perpetuate themselves, most societies depend on
sexual reproduction. The young of the human species cannot survive on their
own, however. To achieve the social good of survival, a society must therefore
institute child-rearing practices. Norms must specified which impose on someone
in society the burden of supporting the young, helping them grow, defending them
against dangers. Who that ends up being is quite varied. Society can specify norms
assigning upbringing responsibilities to those most qualified, for example—certified
child care specialists. Or society can specify norms assigning upbringing
responsibilities to those with the most stake in perpetuation of society—say, senior
members of the child’s social group. Or society can specify norms assign



Argument Has No Function     83

upbringing responsibilities to the biological parents of the child; that at least picks
out definite individuals, and so may lessen disputes. Or... . Further, there are
important normative variations between societies in what the upbringing requires,
how long it takes, where it occurs and so on; within very general limits, all these
variables are up for grabs.

Apply this then to the institution of a norm of argument: the burden of proof.
Arguments which have just been brought to birth need support. To achieve the
social good of argument survival—a necessary precondition for arguments to
achieve their broader social function(s)—a society must therefore institute argument-
rearing practices. Norms must be specified which impose on someone in society
the burden of supporting arguments, helping them grow, defending them against
objections. But who? Society can specify norms assigning argument upbringing
responsibilities to those most qualified—say, an elite corps of argument professionals
in a system of National Argument Clinics, or perhaps the punditocracy. Or society
can specify norms assigning argument upbringing responsibilities to those with
the most at stake in the argument’s growth—say, senior members of the ideological
faction whose position the argument supports. Or society can impose norms
assigning upbringing responsibilities to the person who engendered the argument,
its proponent. Or…. And similar normative alternatives arise in specifying what
objections must be met, how deep must be the response and so on. Within very
general limits, a variety of normative arrangements regarding burden of proof can
promote the survival of an argument, and thus its ability to perform an asserted
function.

The existence of functional alternatives—multiple, yet functional ways of
organizing the joint activity in which arguments arise—means that a single,
determinate set of norms cannot be derived from a function of argument. Now, it
may be that theorists making function claims could defend their derived norms,
not as necessary to achieve the alleged function of argument, but as the most
efficacious or efficient way to do so. In reply I say again: Fine! But show me.
How do you know that imposing on an arguer an obligation to answer just these
objections optimally promotes the mutual understanding, the rational resolution of
disputes, or any other posited social good?

4. There is an alternative to a functional account of the contextual
     norms of argument

Perhaps we would feel compelled to continue to make function claims if they were
the only way to account for the context-dependent norms of argument. But there
are alternatives to function claims in argument theory, as there were to functionalism
in sociology.

The social scientific schools that developed in the 60’s in opposition to
functionalism were diverse, including Goffman’s dramaturgical approach,



84     Jean Goodwin

Garfinkel’s ethnomethodology, Sack’s conversation analysis and Hymes’
ethnography of communication, in addition to the Marxist-inspired conflict theories
noted above. Within this diversity, however, some general tendencies can be
discerned. Anti-functionalist theorists often understood social order as an ongoing
achievement. If something like an argumentative exchange displays regular features,
it is because the participants work hard to make it so. To locate social order, these
theorists therefore focused on the “micro” level of specific interactions, instead of
the “macro” level of an entire society and its institutions. And they explained these
interactions by reference to the choices of individual agents, who act strategically
to achieve their goals.

Consider the following as a representative anecdote; a story told about of Harvey
Sacks and his invention of a new approach to the study of talk:

It was during a long talking walk in the late winter of 1964 that Sacks mentioned
to me a ‘wild’ possibility that had occurred to him. He had previously told me
about a recurrent and much discussed practical problem faced by those who
answered phone calls to the Suicide Prevention Person by suicidal persons
or about them—the problem of getting the callers to give their names. Now
he told me about one call he had seen/heard which began something like this:

A: This is Mr Smith, may I help you.

B:  I can’t hear you.

A: This is Mr Smith [emphasized].

B: Smith.

After which Mr Smith goes on, without getting the caller’s name. And later,
when Mr Smith asks for the caller’s name, the caller resists giving it. On the
one hand, Sacks noted, it appears that if the name is not forthcoming at the
start it may prove problematic to get. On the other hand, overt requests for it
may be resisted. Then he remarked: Is it possible that the caller’s problem in
hearing is a methodical way of avoiding giving one’s name in response to the
other’s having done so? Could talk be organized at that level of detail? And
in so designed a manner? (Sacks, 1995, xvi-xvii).

What if we approached argument in parallel fashion? Is it possible that much of
what we see in argumentative discourse are methodical ways of avoiding, and also
imposing, obligations of various sorts? Could argumentative talk be organized at
that level of detail? And in so designed a manner?

One line of recent scholarship answers “yes” to these questions. Work by Fred
Kauffeld (1995, 1998, 2001, 2002b, 2003), Jean Goodwin (2001a, 2001b, 2002a,
2002b, 2003, 2005a, forthcoming), Beth Manolescu (2005a, 2005b, 2006) and
most recently Bob Pinto (2004) gives accounts of the pragmatic or context-
dependent norms as crafted by agents acting strategically in specific interactions;
we undertake what has been dubbed a design approach (Goodwin, 2002b). The
basic conception of this approach can be summarized in three propositions, standing
in contrast to the three propositions of a function claim.



Argument Has No Function     85

a′ The determinative feature of the context of an argument is the talk in and
through which the argument is made.

 We do not assume that context comes already organized into social forms like
“dialogues” or “critical discussions.” Instead, we take the talk through which, and
within which, arguments are deployed as the primary means by which people
organize a context for their interaction. Because of this, we do not expect to give
a single, uniform account of the norms of argument in all contexts, or even a
cluster of possible accounts, but instead recognize diverse normative environments
created by diverse histories of talk.

b′ A speaker so designs that talk as to create for the recipient(s) reason to
respond as he desires.

We do not take it for granted that argumentative talk is consequential. Quite the
contrary, we observe that it is difficult for speakers to achieve their goals, and we
try to build accounts of how they overcome these difficulties. Specifically, we ask
how a speaker’s uttering some sentences can reasonably be expected to have the
consequences he wants—or indeed any consequences at all—especially from
autonomous recipients who are likely to disagree. We find that speakers use
substantial skill and effort in designing their talk so as to create reasons for the
recipients to respond as desired. To put it in shorthand, instead of assuming that
argumentative talk is functional, we inquire how it is designed to have force.

c′ The norms of argument include those obligations (standards, ideals, etc.)
that his argument must meet (live up to, realize etc.) in order for his talk to have
force.

It is a common strategy for a speaker to create force by creating a local
“normative terrain” in which the response he desires is manifestly obligatory (ideal,
right, prudent etc.). For example, a speaker often would like to oblige his the
recipients of his argument to pay attention to it; then perhaps to seriously consider
it, to respond to it, and eventually to make up their minds about it. As it happens,
strategies which impose obligations on interlocutors commonly require the speaker
himself to commit to producing arguments of a certain quality—that is, to undertake
a burden of proof. In other words, it will often turn out that the same “normative
terrain” that obligates the recipients to respond also obligates the speaker to argue
well. These quality requirements are among the context-dependent or pragmatic
norms of argument.

Now, undoubtedly those of us who put forward “force claims” along the lines
of (a′), (b′) and (c′) will encounter some problems.  I invite the reader to take up
the essays cited above, and start compelling us to address them.

5. Concluding remarks

Some may feel a lingering nostalgia for the critical discussion or other dialogue
types, at least as ideals to be cultivated. And there are undoubtedly a wide range of



86     Jean Goodwin

pre-existing social institutions (speaking broadly) within which arguments can be
found. These may include speech genres (the conference paper), formally organized
occasions (the criminal trial), explicit rule sets (Roberts’ Rules of Order) and
undoubtedly many other more or less enduring patterns of expectations (including
norms) about how argumentative talk will go. What can a design approach say
about these?

As Sally Jackson argued in a seminal Alta paper, even a theorist committed to a
“social-level” analysis of argumentative talk can benefit from the results of an
“individual-level” analysis as well (1992 [1983]). Consider then two ways in which
a committed functionalist might find in the design approach a useful complement
to his own theorizing.

First, argumentative institutions from the conventions of casual politeness up
to full-fledged goal-directed joint activities (if any) are themselves the products of
human ingenuity. They have been created by design to achieve our purposes. We
need to investigate them to account for their designs, to determine (not assume)
their actual consequences, intended and unintended, and to re-engineer them if
necessary to achieve what we want them to achieve. The work of Sally Jackson
(1998, 1999) and Mark Aakhus (2002) is showing the way here.

Second, to function at all, argumentative institutions need the support only
individual argumentative craftsmanship can supply.  Even if one participant knows
that the other understands what a critical discussion (or whatever) ought to be—
that is, even if some institution is “at hand”—getting that other into it may take
substantial discursive work. It is striking, for example, how much force it takes to
get people into child custody mediation—a paradigmatic example of rational dispute
resolution. Participants are under court order, threatened with penalties, and a
third party is present who continually intervenes to push the participants into
compliance with the speech norms. Something similar seems to be the case for
debates between U.S. presidential candidates, which commonly take weeks of
negotiation and hundreds of pages of rules to get started, and which may only take
place because the consequences of appearing to chicken out are so severe. These
are extreme cases, perhaps; but in general, they do suggest that we need to investigate
the strategic resources participants have for making the social institutions or social
ideals for argument effective in an imperfect world. Scott Jacobs (1999, 2000)
and again Mark Aakhus (2003) have been pointing us in this direction.

The functionalism debate in sociology, like scholarly debates generally, faded
out without resolution Indeed, after a time, functionalism revived; now there is a
neofunctionalist school of sociological theorizing that I don’t pretend to understand.

I am calling for nothing more here in argumentation studies. Douglas Walton
has led us in the right direction by turning us to the pragmatic study of argument.
But now that we have proceeded a ways down this path, we need to stop making
naïve function claims. This might mean avoiding function talk entirely—avoiding
throwing out words like “dialogue, cooperation, the purpose of argumentation”



Argument Has No Function     87

and so on, and trying a design approach instead. Or if not this, then let the function
talk become more sophisticated, defending assumptions (a), (b) and (c) against
the objections raised here.7

Notes
1 Roughly: by “joint activity” I mean one in which the participants mutually recognize that their
individual efforts are for the sake of an activity that they share.  A tango is a joint activity;  a war
is not.
2 Functional items are often composed of other functional items and themselves contribute to
larger functions, as mitochondria contribute to cells which contribute to organs which contribute
to the flourishing of the organism. But it is sloppy to speak of “argument having the function of
resolving disputes,” since at best arguments—together with a lot of other talk—contribute to a
joint activity, and it is the joint activity which has that function. The pragma-dialecticians are clear
about this;  some other functionalists aren’t. Still, talk of argument having the function of
disagreement resolution (or other aim), leaving out the joint activity, serves as a convenient
shorthand, and I will occasionally follow it here.
3 What Grice really meant is beyond the scope of this paper, although it should be noted that it is
possible to pursue a Gricean account of the pragmatics of argumentative talk without asserting an
obligation to cooperate; see Kauffeld, 2002a.
4 The controversy is usefully anthologized in Demerath & Peterson (1967) and reviewed by
Abrahamson (1978) and Baert (1998).
5 One response to this problem is to reduce the number of collective goals a given instance of
dialogue is trying to achieve.  Walton suggests this approach for political debate, proposing that
talk in this kind of joint activity should be analyzed as if it were the preferred type of dialogue, a
critical discussion (1998, 224-5). This simplification indeed allows the theorist to derive a single
set of norms, but only at the cost of giving up multi-functionalism, with its empirical flexibility.
6 I take it that something like this objection is the one that Hansen has been addressing to
Johnson’s theory of the “dialectical tier” for the past few years. The obligation to answer
objections seems an “imperfect obligation,” like the obligation of charity:  obliging some kind of
response in general, but not imposing an obligation in specific to meet this objection and not that
one.
7 I thank David Godden, an anonymous reviewer and an audience at the 2005 OSSA conference for
their help in refining the argument of this paper

References

Aakhus, M. (2002). Modeling reconstruction in groupware technology. In F.H. van
Eemeren (Ed.), Advances in pragma-dialectics (pp. 121-136). Amsterdam: Sic Sat.

Aakhus, M. (2003). Neither naïve nor critical reconstruction: Dispute mediators, impasse,
and the design of argumentation. Argumentation, 17, 265-290.

Abrahamson, M. (1978). Functionalism. Englewood Cliffs, NJ: Prentice Hall.

Baert, P. (1998). Social theory in the Twentieth Century. New York: New York University
Press.

Davis, K. (1967). The myth of functional analysis as a special method in sociology and
anthropology. In Demerath & Peterson (Eds.), System, change & conflict (pp. 379-
402). New York: Free Press.



88     Jean Goodwin

Dore, R.P. (1967). Function and cause. In Demerath & Peterson (Eds.), System, change &
conflict (pp. 403-419). New York: Free Press.

Demerath, N.J., & Peterson, R.A. (1967). System, change & conflict. New York: Free
Press.

Elster, J. (1982). Marxism, functionalism, and game theory. Theory and Society, 11, 453-
482.

Eemeren, F.H. van, Grootendorst, R. & Snoeck Henkemans, F. (1996). Pragma-dialectics
and critical discussion. In Fundamentals of argumentation theory: A handbook of
historical backgrounds and contemporary developments (pp. 274-311). Mahwah,
NJ: Lawrence Erlbaum.

Fine, G. (2001). Gifted tongues, Princeton, NJ: Princeton University Press.

Freeman, J. (1991). Dialectics and the macrostructure of argument. Berlin: Foris.

.

Gaskins, R.H. (1992). Burdens of proof in modern discourse. New Haven, CN: Yale
University Press.

Gilbert, M.A. (1997). Coalescent argumentation. Mahwah, NJ: Lawrence Erlbaum.

Godden, D.M. (2005). Deductivism as an interpretive strategy: A reply to Groarke’s
recent defense of reconstructive deductivism. Argumentation and Advocacy, 41,
168-183.

Goodnight, G.T. (1979). The liberal and the conservative presumptions: On political
philosophy and the foundation of public argument. In J. Rhodes & S. Newell (Eds.),
Proceedings of the Summer Conference on Argumentation. Annandale, VA: Speech
Communication

Goodwin, J. (2001a). Cicero’s authority. Philosophy & Rhetoric, 34, 38-60.

Goodwin, J. (2001b) The noncooperative pragmatics of arguing. In E.T. Nemeth (Ed.),
Pragmatics in 2000: Selected papers from the 7th International Pragmatics
Conference (Vol. 2, pp. 263-277). Antwerp: International Pragmatics Association.

Goodwin, J. (2002a). Designing issues. In F.H. van Eemeren & P. Houtlosser (Eds.),
Dialectic and rhetoric: The warp and woof of argumentation analysis (pp. 81-96).
Dordrecht: Kluwer.

Goodwin, J. (2002b). One question, two answers. In H.V. Hansen et al. (Eds.),
Argumentation and its Applications. Windsor: Ontario Society for the Study of
Argumentation, CD-ROM.

Goodwin, J. (2003). Manifestly adequate premises.  In J.A. Blair et al. (Eds.), Informal
Logic @ 25.  Windsor: Ontario Society for the Study of Argumentation, CD-ROM.

Goodwin, J. (2005a). Designing premises. In F.H. van Eemeren & P. Houtlosser (Eds.),
The practice of argumentation (pp. 99-114). Amsterdam: Walter Benjamins.

Goodwin, J. (2005b). What does arguing look like?’ Informal Logic, 25, 79-93.

Goodwin, J. (Forthcoming). Actually existing rules for closing arguments. In F.H. van
Eemeren et al. (Eds.), Proceedings of the Sixth International Conference of the
International Society for the Study of Argumentation. Amsterdam: Sic Sat.

Govier, T. (1987). A new approach to charity. In Problems in argument analysis and
evaluation (pp. 133-158). Dordrecht: Foris.

Jackson, S. (1992 [1983]). The arguer in interpersonal argument: Pros and cons of
individual-level analysis. In W.L. Benoit, D. Hample, & P.J. Benoit (Eds.), Readings



Argument Has No Function     89

in argumentation (pp. 513-521). Berlin: Foris. Originally published in D. Zarefsky,
M.O. Sillars, & J. Rhodes (Eds.), Argument in transition: Proceedings of the third
Summer Conference on Argumentation. Annandale, VA: Speech Communication
Association.

Jackson, S. (1998). Disputation by design. Argumentation, 12, 183-198.

Jackson, S. (1999). The importance of being argumentative: Designing disagreement in
to teaching/learning dialogues. In F. H. van Eemeren, R. Grootendorst, J. A. Blair, &
C. A. Willard (Eds.), Proceedings of the Fourth International Conference of the
International Society for the Study of Argumentation (pp. 392-396). Amsterdam: Sic
Sat.

Jacobs, S. (1999). Argumentation as normative pragmatics. In F. H. van Eemeren, R.
Grootendorst, J. A. Blair, & C. A. Willard (Eds.), Proceedings of the Fourth
International Conference of the International Society for the Study of Argumentation
(pp. 397-403).  Amsterdam: Sic Sat.

Jacobs, S. (2000). Rhetoric and dialectic from the standpoint of normative pragmatics.
Argumentation, 14, 261–286.

Johnson, R. H. (2000). Manifest rationality: A pragmatic theory of argument. Mahwah,
NJ: Lawrence Erlbaum.

Kauffeld, F. J. (1995). The persuasive force of arguments on behalf of proposals. In F. H.
van Eemeren, R. Grootendorst, J. A. Blair, & C. A. Willard (Eds.), Analysis and
evaluation: Proceedings of the Third ISSA Conference on Argumentation (Vol. II,
pp. 79-90). Amsterdam: Sic Sat.

Kauffeld, F. J. (1998). Presumptions and the distribution of argumentative burdens in
acts of proposing and accusing. Argumentation, 12, 245-66.

Kauffeld, F. J. (2001). Argumentation, discourse, and the rationality underlying Grice’s
analysis of utterance-meaning. In E.T. Nemeth (Ed.), Cognition in language use:
Selected Papers from the 7th International Pragmatics Conference (Vol. 1, pp. 149-
163). Antwerp: International Pragmatics Association.

Kauffeld, F.J. (2002a). Grice without the cooperative principle. In H.V. Hansen et al.
(Eds.), Argumentation and its Applications. Windsor: Ontario Society for the Study
of Argumentation, CD-ROM.

Kauffeld, F.J. (2002b). Pivotal issues and norms in rhetorical theories of argumentation.
In F.H. van Eemeren & P. Houtlosser (Eds.), Dialectic and rhetoric: The warp and
woof of argumentation analysis (pp. 97-118). Dordrecht: Kluwer.

Kauffeld, F. J. (2003). The ordinary practice of presuming and presumption, with special
attention to veracity and the burden of proof. In F. H. v. Eemeren, J. A. Blair, C. A.
Willard & A. F. Snoeck Henkemans (Eds.), Anyone with a view: Theoretical
contributions to the study of argumentation (pp. 133-146). Dordrecht: Kluwer .

Lakoff, G. & Johnson, M. (1980). Metaphors we live by. Chicago: University of Chicago
Press.

Manolescu, B. (2005a). Norms of forcibleness. In D. Hitchcock (Ed.), The uses of argument:
Proceedings of a conference at McMaster University (pp. 336-345). Hamilton: Ontario
Society for the Study of Argumentation.

Manolescu, B. (2005b). Norms of presentational force. Argumentation and Advocacy,
41, 139-51.



90     Jean Goodwin

Manolescu, B. (2006). A normative pragmatic perspective on appealing to emotions in
argumentation. Argumentation, 20, 327-343.

Merton, R. (1957). Manifest and latent functions. In Social theory and social structure
(pp. 19-84). Glencoe, IL: The Free Press.

Pinto, R.C. (2004). Reasons, contexts and expectations of rebuttal. A paper presented at
the National Communication Association Convention, Chicago.

Sacks, H. (1995). Lectures on conversation. Malden, MA: Blackwell.

Sanders, L.M. (1997). Against deliberation. Political Theory, 25, 347-377.

Sanders, R.E. (2005). Introduction: LSI as subject matter and as multidisciplinary
confederation. In K.L. Fitch and R.E. Sanders (Eds.), Handbook of language and
social interaction (pp. 1-14). Mahwah, NJ: Lawrence Erlbaum.

Schiffrin, D. (1984). Jewish argument as sociability. Language in Society, 13, 311-335.

Tannen, D. (1998). The argument culture. New York: Random House.

Walton, D. (1992). Plausible argument in everyday conversation. Albany, NY: State
University of New York Press.

Walton, D. (1995). A pragmatic theory of fallacy. Tuscaloosa, AL: University of Alabama
Press.

Walton, D. (1996a). Argument structure: A pragmatic theory. Toronto: University of
Toronto Press.

Walton, D. (1996b). Argumentation schemes for presumptive reasoning. Mahwah, NJ:
Lawrence Erlbaum.

Walton, D. (1998). The new dialectic: Conversational contexts of argument. Toronto:
University of Toronto Press.

Walton, D. (2006). Fundamentals of critical argumentation. Cambridge: Cambridge
University Press.

Walton, D. & Krabbe, E.C.W. (1995). Commitment in dialogue: Basic concepts of
interpersonal reasoning. Albany, NY: State University of New York Press.

Young, I.M. (1996). Communication and the other: Beyond deliberative democracy. In S.
Benhabib (Ed.), Democracy and difference: Contesting the boundary of the political
(pp. 120-135). Princeton, NJ: Princeton University Press.

Jean Goodwin
Department of English

223  Ross Hall
Iowa State University

Ames, Iowa 50011
U..S.A.

goodwin@iastate.edu