Hoffmann2018 Article TheElusiveNotionOfArgumentQual

Argumentation (2018) 32:213–240
https://doi.org/10.1007/s10503-017-9442-x
The Elusive Notion of “Argument Quality”
Michael H. G. Hoffmann1
Published online: 17 November 2017

© Springer Science+Business Media B.V., part of Springer Nature 2017
Abstract We all seem to have a sense of what good and bad arguments are, and
there is a long history—focusing on fallacies—of trying to provide objective
standards that would allow a clear separation of good and bad arguments. This
contribution discusses the limits of attempts to determine the quality of arguments.
It begins with defining bad arguments as those that deviate from an established
standard of good arguments. Since there are different conceptualizations of “argu-
ment”—as controversy, as debate, and as justification—and since arguments in each
of these senses can be used for different purposes, a first problem is that we would
need a large variety of standards for “good” arguments. After this, the contribution
focuses in particular on proposals made in the literature on how to assess the quality
of arguments in the sense of justification. It distinguishes three problems of
assessment: How to determine (1) whether reasons are acceptable, (2) whether
reasons are sufficient to justify the conclusion, and (3) how to identify arguments in
real-world speech acts and texts? It is argued that limitations of argument assess-
ment result from unavoidable relativism: The assessment of many—if not most—
arguments depends on the epistemic situation of the evaluator.
Keywords Argument assessment · Argument definitions · Argument quality ·

Background knowledge · Epistemic constraints · Fallacies · Good arguments ·
Standards
& Michael H. G. Hoffmann

[email protected]
1
School of Public Policy, Georgia Institute of Technology, 685 Cherry Street, N.W., DM Smith
Building, Atlanta, GA 30332-0345, USA
123
214 M. H. G. Hoffmann
1 Introduction
Everybody who teaches critical thinking or anything that requires the construction
of arguments should have sufficiently clear ideas about how to assess the quality of
arguments. This is even more important for research. If we want to know whether
people are able to create arguments, objective standards for the evaluation of
arguments are needed. And it is certainly important to train students in argument
assessment so that educated citizens know what is worthwhile to consider and what
is not in the often highly intense conflicts we are facing today.
The need for fostering the ability to distinguish between good and bad arguments
is, of course, well known since Aristotle started to determine criteria for separating
true “syllogisms” (οἱ συλλογισμοί) from those that “only seem to be, but are not
really” syllogisms, as he wrote in the beginning of his Sophistical Refutations
(Aristotle Soph. el. [Forster] 164a24). He calls the fake syllogisms “paralogisms”
(παραλογισμοί, a21), literally “besides, against, or deviating from a logismos.”
Paralogism, together with “sophism,” are the Greek terms that were later translated
as fallaciae in Latin (from fallax, “deceitful, deceptive, fallacious”).
Classic Greek does not have a specific term for “argument,” but there are two
concepts that anticipate the idea. The first one is logos (λόγος) which has a very
large number of meanings, ranging from computation, reckoning, and account to
relation, proportion, rule, and law, to reasoning and thinking, and to explanation,
ground, statement of a theory, discussion, debate, speech, narrative, reason as a
faculty, and any verbal expression or utterance.1 Among these meanings of logos we
find “having logos” in the sense of “being arguable” or “reasonable” (Plato Phd.,
62d) and in the sense of “having a justification,” as in Plato’s Theaetetus when
Socrates defines knowledge as “true belief with logos” (Plato Tht., 201c-d). Derived
from logos we find also logismos which meant originally “counting, calculating,
reckoning,” but later also “reasoning, reason, and reasoning power.”2
The second Greek term that anticipates “argument” is apodeixis (ἀπόδειξις)
which means “showing forth, making known, demonstration, and proof.” Aristotle’s
Prior Analytics is dedicated to the “science of apodeixis” and the act of
“apodechein” (Aristotle An. Pr. 24a11); or—as Stephen Toulmin translates
apodechein to determine what his famous book The uses of argument is all about
—“the way in which conclusions are to be established” (Toulmin 2003 [1958], p. 2).
The Prior Analytics are the books in which Aristotle lays the foundations of
syllogistic logic and where he defines “syllogism” as a “logos in which from certain
assumptions something different follows with necessity,” merely based on these
assumptions without the need of anything else (24b19-24).
Since in Aristotle the terms paralogism and syllogism are so closely related, it
seems appropriate to characterize as “fallacies” in Aristotle’s writing every form of
inference that deviates, in one way or another, from valid syllogisms. This
1
See http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0057%3Aentry%
3Dlo%2Fgos.
2
See http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0057%3Aalphabetic
+letter%3D*l%3Aentry+group%3D46%3Aentry%3Dlogismo%2Fs.
123
The Elusive Notion of “Argument Quality” 215
understanding of fallacies is reflected in Hamblin’s definition when he writes in his

standard work on Fallacies: “A fallacious argument, as almost every account from
Aristotle onwards tells you, is one that seems to be valid but is not so” (Hamblin
1970, p. 12). Hamblin’s book is well-known as the first large attempt to summarize
the history of “bad arguments” since Aristotle and to give research on argument
assessment or appraisal a more systematic direction—countering thus the centuries-
old practice of just providing lists and examples of fallacies as we can still find them
today in many introductory books on logic and critical thinking. However, it seems
that almost all research on argument quality after Hamblin rejects his definition of
fallacy (Hansen 2015). Whereas this definition seems perfectly appropriate for
characterizing what “fallacy” means for Aristotle, a possible generalization of what
we discussed so far could be: a fallacy is any deviation from a certain standard of
good arguments. This way, what counts as fallacious depends on such a standard.
This idea is reflected in the way pragma-dialectics defines fallacies. Referring to the
15 “rules for a critical discussion” that form the core of van Eemeren and
Grootendorst’s Systematic Theory of Argumentation, they count as fallacy: “Every
violation of any of the rules of the discussion procedure for conducting a critical
discussion” (van Eemeren and Grootendorst 2004, p. 175).
Since the term “fallacy,” in common usage, is still very much connected with
various lists of unacceptable argument patterns, and since, as Hans V. Hansen puts it
in his encyclopedia article on the subject, the “kinds of mistakes one can make in
reasoning are generally thought to be beyond enumeration” (Hansen 2015), I prefer
to talk about “bad arguments” instead of “fallacies.” If we define bad arguments as
arguments that deviate from an established standard of good arguments, we face
immediately a series of questions: Is there—or: can there be—a singular standard of
good arguments and, if so, what is it? If not: What are generally accepted standards
of good arguments?
Looking at the literature, we see a multitude of standards. We mentioned already
Aristotle’s syllogisms and the “rules for a critical discussion” in pragma-dialectics.
But there are more. In an early attempt to overcome the limitations of using formal
or deductive logic as the only standard for good arguments, Ralph Johnson and
Anthony Blair proposed in Logical Self-Defense three criteria to assess the quality of
arguments which are now widely known as the ARS criteria. The premises that
support the conclusion of an argument should be acceptable, relevant, and sufficient.
As they write in the Preface to the textbook’s 2nd edition: “These criteria have the
advantage of including deductive validity and inductive strength as special cases,
while at the same time they leave open the possibility that there are other legitimate
kinds of inference in arguments besides valid deductions and strong inductions”
(Johnson and Blair 2006 [1977], p. xiii).
Those “other legitimate kinds of inference in arguments” are also the focus of the
literature on argumentation schemes which provides yet another standard of good
arguments. In Argumentation Schemes, Douglas Walton, Chris Reed, and Fabricio
Macagno provide the currently most comprehensive compendium of these schemes
(Walton et al. 2008). For each argumentation scheme, they formulate a specific set
of “critical questions” that are designed so that a respondent to an argument can
123
critically assess its quality by focusing on features that are specific for a particular
scheme (pp. 3, 9).
These four examples of work that attempts to establish standards for good
arguments should be sufficient to show that a multitude of standards has been
proposed. This fact in itself does not exclude the possibility that all these and other
standards could be combined in a singular framework. As already mentioned,
according to Johnson and Blair the ARS criteria can include the syllogistic standard
as a special case, and so it might also be possible to consider critical questions as
specifications of these three criteria. However, any such unification of argument
standards faces one important problem. It is far from clear what an “argument” is
when we talk about “good arguments.” Whereas Walton, Reed, and Macagno still
present many of their “argumentation” schemes according to the Aristotelian pattern
of “major premise,” “minor premise,” and “conclusion,” for van Eemeren and
Grootendorst an “argumentation” is something much larger: “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 refuting
the proposition expressed in the standpoint” (van Eemeren and Grootendorst 2004,
p. 1). Their 15 rules for a critical discussion determine “a methodological regulation
of argumentative discourse and texts. Together, the rules combine to constitute a
dialectical discussion procedure” (p. 57). This procedure is designed to resolve “a
difference of opinion” over four “dialectical stages”:
The confrontation stage in which the difference of opinion is developed, the
opening stage in which the procedural and the other starting points are
established, the argumentation stage in which the argumentation is put forward
and subjected to critical reaction, and the concluding stage in which the
outcome of the discussion is determined (van Eemeren and Grootendorst
2004, p. 134).
Obviously, an “argumentative discourse” differs fundamentally from an “argument”
in the sense of a few propositions that include just premises and a conclusion. Since
the standards for good arguments depend on what is meant by the term “argument,”
I will first—in the following, second section of this paper—distinguish three very
different conceptualizations of the term “argument.” After that, I am going to
discuss for each of them a variety of standards that have been proposed for
distinguishing good and bad arguments. This discussion will show that these
evaluation criteria face problems that make it unlikely that we ever will be able to
define an objective standard for good “arguments” that would allow a clear
distinction from bad ones. This conclusion of my arguments is much more
pessimistic than most of the positions discussed.
2 Quality of What?
Since reflection on arguments and their role is important in many disciplines, it

might not be surprising that the concept itself—argument—has been used in a
variety of ways. To avoid confusion and misunderstanding, clear definitions are
123
needed. However, since every definition excludes those usages of a term that are not
covered by a proposed definition, definitions always constrain fields of deliberation.
For example, if we define “argument” following Aristotle’s use of apodeixis as
syllogism, only those contributions to a discussion of “argument quality” will be
considered legitimate that define “argument” in the same way. Definio, which
derives from finis, “end, boundary, limit,” means literally “drawing a boundary
around something, delineating it.” A definition determines what is inside and what is
outside. It is, thus, a question of responsibility and fairness first to survey an area of
concept usages before one determines who is in and who is out.
Looking also beyond the more narrow field of argument theory and informal
logic, it seems to me that at least three fundamentally different usages of the term
“argument” should be clearly distinguished: argument as controversy between two
positions; argument as debate; and argument as justification. Since argumentation
skills play a large role in science education and in education in general, the
following overview of literature that defines argument in one of these three ways
includes publications in these areas. Additionally, I will talk about their historical
origins and also about software tools in which they are implemented. Computer-
supported argument visualization (CSAV) plays an increasing role in education,3
but most CSAV tools realize just one of the three conceptualizations of argument.4
A clear differentiation of the various definitions of argument is, thus, especially
important to get a better understanding for what these tools can be used, and of their
limitations.
2.1 Argument as Controversy Between Two Positions
When van Eemeren et al. (2014) write in the newest edition of the Handbook of
Argumentation Theory: “a communicative activity that is not aimed at resolving a
difference of opinion is not considered as argumentation” (p. 2, Fn. 1), they limit—
if we assume that they are using the concepts “argument” and “argumentation”
interchangeable5—the use of argument to controversies between two positions.
There are exactly two roles in the pragma-dialectical approach to argument: the
“protagonist” and the “antagonist.” In the same vein, we find definitions of
argument “as expansion of conversation around disagreement” (Jackson 2015,
p. 247); as “any form of collaborative activity that involves confronting cognitions
and their foundations” (Andriessen et al. 2003, p. 2); and as “messages and
discussions, in which subjective perspectives or differences of opinion are important
and relevant” (Schneider et al. 2013, p. 160). Historically, this first interpretation of
argument in the sense of controversy can be traced back to the sophist Protagoras (c.
490—c. 420 BC) who, according to Diogenes Laertius, “was the first one who
claimed that there are two opposing logoi with regard to everything” (D.L. IX 51).
3
See Kirschner et al. (2003), Andriessen et al. (2003) and Scheuer et al. (2010, 2013).
4
An exception is Rationale which offers templates for each of them: https://www.rationaleonline.com.
5
Evidence for this can be seen in the fact that the index in van Eemeren and Grootendorst 2004 lists only
“argument(ation),” and not both concepts separately.
123
Some of the definitions of argument that point to controversy include additionally

the idea of justification. Deanne Kuhn, for example, defines what she calls a
“dialogical argument” in contrast to a “rhetorical” one as an argument that
entails juxtaposition of two opposing assertions. Most often, we think of an
argument as a dialogue between two people who hold opposing views. Each
offers justification for his or her own view; in addition (at least in a skilled
argument), each rebuts the other’s view by means of counterargument. (Kuhn
1991, p. 12)
Similarly, whereas pragma-dialecticeans define “argumentation,” as we saw, as a
procedure aimed at resolving a difference of opinion that unfolds over four stages,
they sometimes use the term “argument” in the sense of “reason(s) for a standpoint”
(van Eemeren and Grootendorst 2004, p. 61; but they use for the latter also
“argumentation,” see pp. 106–107). We find this distinction between “argumenta-
tion” and “argument” also in Barth and Krabbe’s “theory of rational and critical
argumentation” which forms one of the foundations of pragma-dialectics.6
Whereas all these approaches combine, thus, argument as controversy and
argument as justification, we find in Dung’s abstract argumentation framework,
which is widely adopted in the computational literature on argumentation, an
understanding of argument which limits the meaning of the concept exclusively to
an “attack relationship.” But it is not this attack relationship or a controversy which
is called “argument” but the two propositions which are related in form of an attack.
An argument, writes Dung, “is an abstract entity whose role is solely determined by
its relations to other arguments. No special attention is paid to the internal structure
of the arguments” (Dung 1995, p. 326). The only “relation” that is relevant for
Dung’s definition of argument is “a binary relation representing the attack
relationship between arguments” (pp. 325–326). Thus, in Dung’s framework “each
argument is regarded as atomic. There is no internal structure to an argument”
(Besnard et al. 2014, p. 1).
Among computer-supported argument visualization (CSAV) tools, Consider.it,
ProCon.org, Debate.org, and LivingVote are specifically designed to visualize
arguments in the sense of controversy between two positions.7 All of them,
however, provide also functionality to present arguments in the sense of
justification.
2.2 Argument as Debate on an Issue
Recent work on this second usage of “argument,” especially in the area of CSAV, is
heavily influenced by Rittel and Webber’s seminal paper on “wicked problems”
(Rittel and Webber 1973) and Horst Rittel’s conceptualization of “Issue-Based
Information Systems” (IBIS) as a way to approach these problems (Rittel and Noble
6
See Barth and Krabbe (1982, p. 25 and p. 52) regarding “argumentation,” and pp. 15–18 for
“argument”; see also their notion “chain of arguments” on p. 63 and others. It is a bit surprising that
neither van Eemeren and Grootendorst nor Barth and Krabbe define “argument.”
7
See https://consider.it, http://www.procon.org, http://www.debate.org/, and http://www.livingvote.org/.
All accessed February 15, 2017.
123
1989). Rittel and Webber characterized “wicked problems,” in contrast to “tame

problems,” among other things as problems whose understanding depends on
someone’s point of view. “There is no definitive formulation of a wicked problem.”
“The formulation of a wicked problem is the problem!” (Rittel and Webber 1973,
p. 161).
Rittel and Webber recommended that the multi-perspectivity of wicked problems
should be addressed based on “an argumentative process in the course of which an
image of the problem and of the solution emerges gradually among the participants,
as a product of incessant judgment, subjected to critical argument” (Rittel and
Webber 1973, p. 162). Since the focus is here on “an open-ended, dialectical process
of collaboratively defining and debating issues” (Buckingham Shum 2003, pp. 12–
13), many scholars—most prominently in Visualizing Argumentation, the volume in
which the term “computer-supported argument visualization” (CSAV) has been
coined (Kirschner et al. 2003)—refer to an increasing number of activities when
they talk about “argument visualization.” Conklin’s Compendium software—which
is closest to Rittel’s original IBIS approach—allows the user to put questions,
answers/ideas, pros and cons, and decisions on an “argument” map (Conklin 2006,
pp. 88–90). Others went beyond that and added things such as opinions, predictions,
hypotheses, theories, data, frameworks, and evaluation criteria; and they allowed to
express doubt and disbelief; and to reify, contrast, criticize, and integrate
perspectives (Kirschner et al. 2003, pp. vii, 12, 52; Benn and Macintosh 2012,
p. 77). This way, everything that people contribute in a debate on an “issue” can be
counted as part of an argument.
The fact that Kirschner et al. (2003) have indeed such a broad concept of
“argument” in mind—something like “a set of elements that are contributed in a
debate on an issue”—is evidenced by the way they derive the meaning of
“argument” from the Latin argumentum and arguere and their root in the Greek
adjective ἀργός (“shining, glistening”). They write:
The etymology lies in the Latin argumentum, from arguere which means to
make clear. And this is what it’s all about. How do we make clear – at least so
far as one is willing and able – what we think, what we mean, what we believe
and need, so that we can work together to define and solve the problems that
confront us? The above definitions [from the American Heritage Dictionary,
4th ed.] frame argumentation not only as discourse for persuasion, logical
proof, and evidence-based belief, but more generally, discussion in which
disagreements and reasoning are presented. (Kirschner et al. 2003, p. ix)
Another, more recent approach to argument in the sense of debate has been
developed by Mark Aakhus and Marcin Lewinski (Lewinski and Aakhus 2014;
Lewiński 2015; Aakhus and Lewiński 2017). The goal of their “polylogical
approach” is to extend argumentation theory beyond its well-established focus on
dialectical or dialogical structures because: “argument in contemporary controver-
sies and deliberation involves many players, many positions, and many places”
(Aakhus and Lewiński 2017, p. 179). They define an “argumentative polylogue” as
a discussion among “multiple different argumentative parties defending their
distinct positions” (p. 181).
123
This second definition is broader than the first one, argument as controversy
between two positions. When it comes to wicked problems, a large number of
stakeholders with different points of view may be involved. Since they all should
have a say in debating the problem, an “argument as debate” can be a very large
item. Klein et al. (2012), for instance, talk about “the largest single argument map
ever created”—that means: one argument—when they report on a debate in which
220 participants used the CSAV tool Deliberatorium to post over 3000 issues, ideas,
and pros and cons, in addition to 1900 comments, over a period of 3 weeks.
Software tools that realize such an understanding of argument as debate include
Evidence Hub, Rationale Online, bCisive online, Mind Meister, and Cope_it which
all present “arguments” in graphical form. The Deliberatorium represents them in
the form of threads.8
2.3 Argument as Justification of, or Support for a Claim
This third definition of “argument” is widely shared in philosophy. The set of

propositions that is involved in an argument as justification might be larger—as in
Toulmin’s model of argument which includes a claim, data (or grounds), warrant,
backing, qualifier, and rebuttal9—or smaller, as in Trudy Govier’s definition: “An
argument is a set of claims in which one or more of them—the premises—are put
forward so as to offer reasons for another claim, the conclusion.”10 As James
Freeman writes: “In any argument, there must be at least one premise and at least
one conclusion” (Freeman 2011, p. 3). We can find an early definition of argument
as justification with the rhetorician Quintilian:
an argument is a process of reasoning which provides proof and enables one
thing to be inferred from another and confirms facts which are uncertain by
reference to facts which are certain. (argumentum est ratio probationem
praestans, qua colligitur aliquid per aliud, et quae, quod est dubium, per id
quod dubium nonest, confirmat, Quintilian 1921, 5 10.11, around 95 AD)
Whereas Quintilian understands “argument” here as a process, more common seems
to be to see it as a product. Early evidence for this view is given in a report by
Diogenes Laertius on the philosophy of the Stoics (second century BC):
8
See http://evidence-hub.net/, https://www.rationaleonline.com/, https://www.bcisiveonline.com/,
https://www.mindmeister.com, http://copeit.cti.gr/, and http://cci.mit.edu/klein/deliberatorium.html. All
accessed February 15, 2017.
9
Toulmin 2003 [1958]. For Toulmin, the corresponding model constitutes a “micro-argument” which he
contrasts to the “macro-argument” of which it is a part like an organ in an organism. Also macro-
arguments are considered exclusively as justifications. Even though they might proceed over a number of
pages or hours, the lead “to the final presentation of a conclusion” (p. 87). See also Freeman (2011) and
Benn and Macintosh (2012). It should be noted that Freeman interprets Toulmin’s model as “analyzing
arguments as dialogical exchanges between a proponent and a challenger” (p. 12), that is, as representing
a controversy.
10
Govier (2010, p. 1). See also Hamblin (1970, ch. 7), Johnson and Blair (2006 [1977], pp. 9–10), Fisher
(2001, p. 235), and Groarke et al. (2008, pp. 1–9). The fact that many of these authors discuss arguments
as situated in the context of dialogues, controversies, or debates should not distract from the fact that their
definitions of the term argument refer only to a certain set of propositions.
123
Most useful, they say, is the theory of the syllogisms because it brings the
demonstration to light, which contributes much to the correction of teachings,
and because order and memory bring scientific understanding to light. The
logos itself would be a composition of premises and conclusion (Εἶναι δὲ τὸν
λόγον αὐτὸν σύστημα ἐκ λημμάτων καὶ ἐπιϕορᾶς); the syllogism an inferential
logos from those; and apodeixis a logos that reveals the less understood from
the better understood.11
Sextus Empiricus (c. 160–210 AD) uses the same definition which we can take as
indication that by his time the use of “logos” in the sense of “argument as
justification” was well-established:
A logos is a composition of premises and conclusion. We call “premises” the
axioms that are assumed, based on agreement, for the establishment of the
conclusion, and “conclusion” the axioms that are established based on the
premises.” (Sextus 1912, II 135-6. For more see Bochenski 1970, §21)
A summary and discussion of the modern view of argument as justification has been
provided by David Hitchcock (Hitchcock 2007). Based on Hitchcock’s discussion of a
large variety of argument definitions, his own definition, and the criticism of Hitchcock’s
definition by Geoff Goddu (2009), I suggested elsewhere to define an argument as “a
premise-conclusion sequence so that either one or more premises are intended to support a
conclusion or a conclusion is intended to be justified by one or more premises” (Hoffmann
2016a, with further discussion). Examples of CSAV tools that focus on justification are
OVA, Rationale Online, bCisive online, Argunet, and AGORA-net.12
2.4 More Defintions of Argument
These three conceptualizations of argument—as justification, as controversy

between two positions, and as debate on an issue—can be seen as prototypes of
very different usages of the concept. To define them as such we can determine for
each conditions that are both necessary and sufficient:
● For an argument in the sense of justification it is both necessary and sufficient

that there is a set of propositions among which one is the conclusion and at least
one other a reason.
● For an argument in the sense of controversy between two positions it is both
necessary and sufficient that there are two conflicting positions on the same
proposition (pro and con).
● For an argument in the sense of debate it is both necessary and sufficient that
there are more than two people with conflicting positions who communicate
about the same issue.
11
Lives of Eminent Philosophers (VII 45). My translation from http://www.perseus.tufts.edu/hopper/
text?doc=Perseus%3Atext%3A1999.01.0257%3Abook%3D7%3Achapter%3D1.
12
See http://ova.arg-tech.org/, https://www.rationaleonline.com/, https://www.bcisiveonline.com/, http://
www.argunet.org/, http://agora.gatech.edu/. All accessed February 15, 2017.
123
In many theories these idealized or pure forms of argument are combined, as in

Barth and Krabbe’s dialogical approach to logic where “argumentation” as
controversy and “argument” as justification are distinguished, or in IBIS-based
debate systems which always include the visualization of “pro” and “con” relations.
Other theories provide alternative categorizations. Hamblin distinguished argument
as justification from “dialectical systems” in which “arguments are put forward”
(Hamblin 1970, ch. 7 vs. ch. 8; p. 254). Similarly, Ralph Johnson defined arguments
as having both an “illative core” at which they are understood as “reasons presented
in support of a conclusion, or as ‘premises’ leading to a conclusion,” and a
“dialectical tier.” This dialectical tier is conceptualized as “the second level of an
argument,” on top of justification as its first level—a “layer of argument in which
the arguer discharges his or her dialectical obligations by anticipating and
responding to objections, criticism, and so on” (Johnson 2009a, pp. 55, 64; Johnson
2000; see also Freeman 2011).
In the literature on “computational argumentation” we find a distinction between
“rhetorical,” “dialogical,” and “monological models” of argument. In the last one,
“argument” is understood in the sense of justification and as product, whereas it is
considered a process in dialogical models (Bentahar et al. 2010; see also Johnson
2000). “Rhetorical models” are those that take into account how an argument—
obviously in the sense of justification—is perceived by an audience and how
arguments are used “as a means of persuasion” (Bentahar et al. 2010, p. 214; see
also Perelman and Olbrechts-Tyteca 1969 [1958]). Another example from the
literature on computational models of argument is work on “structured argumen-
tation.” Here a distinction between “argument” as justification and “attack” as “a
binary relation over arguments that denotes when one argument is in conflict with
another argument” has been proposed (Besnard et al. 2014, p. 2).
It should be recommendable—to avoid confusion and misunderstanding—to use
the term “argument” only after making explicit which necessary and sufficient
conditions are required for a particular usage of the term. For the IBIS approach, for
example, a definition of argument could say that it is both necessary and sufficient
that there are more than two people with conflicting positions who communicate
about the same issue and who provide reasons for or against particular claims; for
the pragma-dialectical approach, a definition could say that it is both necessary and
sufficient that there are two conflicting positions on the same proposition and that
each of them is or gets justified by reasons or refuted by objections; and so on.
Defining different usages of “argument” along these lines takes arguments only
as “products” into account, that is as something that is completed and can be studied
in form of a representation, be it a text, a graphical argument map, or a recording. Of
course, all three forms of argument that I distinguished above can be realized both
as product and as process; we can find each of them materialized in a representation,
as well as unfolding in time, for example in dialogues, monologues, or in individual
or collaborative argument mapping. It is important to note that it makes a difference
whether we assess an argument—in any of the three forms—as a product or as a
process. Both will require different standards. Whereas, for example, an argument
as justification which is studied as a process might be assessed with regard to the
123
efficiency of reaching a certain outcome, this criterion will not play any role when
only the outcome or product is studied. This contribution focuses mainly on
argument as product.
It seems to me that many “rhetorical arguments” can be subsumed under
argument as controversy because the only relevant difference that I can see in the
literature mentioned above is whether the audience is explicitly taken into account
or not (for a more differentiated look at rhetorical arguments see Zarefsky 2016).
Moreover, it should be noted that the exclusive focus on persuasion in many
rhetorical models of argument points to a problem that needs to be considered in its
own right: What is the function or purpose of arguments? It seems to me that this
question should be separated from questions of definition. There is no need to
restrict any of the three proposed conceptualizations of argument to a specific
purpose. As we will see, whatever definition of argument one prefers, arguments can
be used for all sorts of purposes.
3 Three Forms of Argument: Three Standards of Quality?
Just as the quality criteria of a hammer will differ from those of a good pudding, it is
necessary to distinguish various standards of argument quality according to what is
meant by the term “argument.” A first point that should be obvious is that lists of
criteria for the three basic conceptualizations of argument we distinguished—
argument as controversy, debate, and justification—will vary according to the
complexity of what is meant by “argument.” Whereas an argument in the sense of
justification can consist of just a few propositions, an argument as controversy
requires two parties that engage in a set of activities which will minimally include
the proposal of a position or statement and a critical response to this proposal; more
often, however, it will entail arguments in the sense of justification, attempts to
refute or counter these arguments, and may be—as in pragma-dialectics—even an
entire procedure that unfolds over several well-defined steps. In arguments in the
sense of debate, finally, the number of parties can be very large and—if we take into
account that this approach has originally been developed to deal with wicked
problems—the argument can be very complex in so far as wicked problems are
problems in which not even the formulation of the problem is uncontroversial (Rittel
and Webber 1973, p. 161: “The formulation of a wicked problem is the problem!”).
Starting with the quality of debates, we can distinguish two strategies. One has
been proposed by Marcin Lewiński who takes the normative standards developed
for dialectical approaches to argument (argument as controversy, in my terminol-
ogy) and asks whether there can be similar standards of rationality in argumentative
polylogues—a task that reveals many problems, as he shows (Lewiński 2015). A
second strategy focuses on a long history of formulating rules for debates in which
many different standards have been proposed (e.g., Johnson 2009b; Wodak and
Koller 2010). The best known standard may be Robert’s rules of order revised for
deliberative assemblies (Robert 1915; Robert and Robert 2011). In order to show
what kinds of problems a determination of criteria for good debates poses, let me
briefly discuss two examples. An interesting list of those criteria is given in the
123
Discourse Quality Index (DQI) which has originally been developed to assess the
quality of parliamentary debates (Bächtiger 2005; Steiner et al. 2004). It focuses on
items such as interrupting another speaker, respectful language, respectful listening,
respect toward counterarguments, the “level of justification of arguments” (a
speaker presents no argument, one, two, or more with or without “linkage”), the
reference to abstract principles, and “constructive politics” such as indicating a
change in position and the acknowledgment of values hold by other speakers
(Caluwaerts and Deschouwer 2014, p. 437; Steiner 2012).
A second very interesting standard to assess the quality of debates evolved on the
internet platform ChangeMyView (CMV). CMV is a “subreddit for people with an
opinion that they accept may be flawed, and who want help in understanding other
perspectives on the issue.”13 The name “change my view” is meant as a request:
Tell me something that I did not consider so that I can change my position. The
platform defines a standard of good debate by five “submission rules” and five
“comment rules”14:
Submission rules for “original posters” (OP):
A. Explain the reasoning behind your view, not just what that view is
(500 + characters required).”
B. “You must personally hold the view and be open to it changing.”
C. “Submission titles must adequately sum up your view and include ‘CMV:’ at
the beginning.”
D. “No ‘meta posts.’ Please post to/r/ideasforcmv instead.”
E. “Only post if you are willing to have a conversation with those who reply to
you, and are available to start doing so within 3 h of posting.”
Comment rules:
1. “Direct responses to a CMV post must challenge at least one aspect of OP’s
current view (however minor), unless they are asking a clarifying question.
Arguments in favor of the view OP is willing to change must be restricted to
replies to comments.”
2. “Don’t be rude or hostile to other users. Your comment will be removed even if
the rest of it is solid.”
3. “Refrain from accusing OP or anyone else of being unwilling to change their
view. If you are unsure whether someone is genuine, ask clarifying questions. If
you think they are exhibiting un-CMVish behavior, please message the
mods.”15
4. “Award a delta if you’ve acknowledged a change in your view. Do not use
deltas for any other purpose. You must also include an explanation of the
13
https://www.reddit.com/r/changemyview/. Accessed February 17, 2017.
14
https://www.reddit.com/r/changemyview/wiki/rules. Accessed February 17, 2017. On this web page,
the interested reader can find extensive justifications and clarifications for each of these rules. It has to be
noted that formulations that I am quoting here have changed substantially over the past years.
15
“Mods” are the moderators. CMV is moderated by volunteers.
123
change along with the delta. Delta abuse includes sarcastic deltas, joke deltas,
super-upvote deltas, etc.”
5. “No ‘low effort’ posts. This includes comments that are only jokes or ‘written
upvotes.’ Humor and affirmations of agreement contained within more
substantial comments are still allowed.”
These rules are visible on the main page and they are explained, justified, and
discussed with a lot more detail on a separate rule page. The following quote,
referring to the first submission rule, shows the level of detail:
Your explanation should not try to convince others of your view by arguing
for it. It should explain why you hold the view. This is a subtle difference. For
example: “I believe this because X, Y, Z” is much better than “X is good
evidence for this, Y reinforces that, and Z clinches why it must be true”. Try to
speak in terms of what you believe, rather than by presenting facts and
citations. Long explanations are fine, if they are necessary in order to explain
what your view is or why you have come to believe it. However, excessively
long supporting arguments for your view are a warning sign that you’re trying
to convince others.16
It might be surprising to see that users of CMV are not supposed to argue for their
view. In spite of the fact that CMV can be counted as a debate platform, the purpose
of these debates is explicitly not to convince other people. Instead, CMV is designed
—just as its name suggest—to change people’s own views by means of debate.
The important point that becomes visible with the two sets of rules quoted above
is that a standard of good argument is not only determined by what is meant by the
term “argument,” but it also depends on the purpose a specific type of argument is
supposed to fulfill. To convince opponents in a parliamentary debate is a different
purpose than inviting others to get one’s own view changed. The fact, however, that
arguments can be used for all sorts of purposes means that any list of quality criteria
will vary according to the purpose an argument is supposed to serve. Since purposes
can be divided, it seems, ad infinitum or specified with ever more detail, there is
hardly any hope that there can ever be a complete list of criteria that could be used
to assess the quality of every possible argument in the sense of debate. There is no
limit to the kind of subtlety mentioned in the quote above about one of the CMV
rules.
The assumption that quality criteria need to vary according to the purpose an
argument is supposed to fulfill, as well as the thesis that purposes can be
distinguished ad infinitum, finds further support when we turn to argument as
controversy. We saw already that for pragma-dialectics the purpose of argumen-
tation is “resolving a difference of opinion on the acceptability of one or more
standpoints by means of a critical discussion” (van Eemeren and Grootendorst 2004,
p. 132). The 15 rules for a critical discussion van Eemeren and Grootendorst
propose can be used to determine the quality of an argumentation in this dialectical
sense. However, a difference of opinion cannot only be resolved “when the
16
https://www.reddit.com/r/changemyview/wiki/rules. Accessed February 17, 2017.
123
arguments advanced lead the antagonist to accept the standpoint defended, or when
the protagonist retracts his standpoint as a consequence of the critical reactions of
the antagonist” (p. 133; see also Rule 14 on p. 154); a difference of opinion can also
be resolved when both parties give up their original standpoint and agree on a new
position. Whereas “winning an argument” is certainly the goal in a court case where
the accused is either guilty or not, in most conflicts it is usually a bad idea to divide
the world into winners and losers because, as it is well-known in conflict research,
“disgruntled losers can cause problems later on by challenging the decision.”17
The criteria we would use to assess the quality of processes whose purpose is
either to win an argument or to build a compromise as a new position will differ. If
we focus on procedural rules as standards of quality—and not just on achieving
agreement as the only criterion of argument quality18—the goal of formulating a
higher-level consensus position will probably require specific rules or procedural
steps that are designed to motivate parties to give up or at least modify their original
position. Again, depending on how exactly the purpose of an argument is defined,
quality standards will differ. And since there is no end in sight regarding possible
distinctions of purposes, determining criteria for good arguments in the sense of
controversy will be as challenging as we saw it in the discussion on argument as
debate.
Things should be different, however, when we consider argument as justification
because this conceptualization includes already its purpose, and it is just one:
justification. It is clear that arguments in the sense of justification can be used for all
sorts of further purposes beyond justifying or supporting a claim by reasons,19 but
here—in contrast to argument as debate and argument as controversy—it is
appropriate to distinguish a primary purpose from a large, and maybe infinite set of
secondary purposes. However, as we will see in the next section, even this most
simple understanding of an argument—just a few statements in a certain
relationship—has its problems when it comes to quality criteria.
4 Determining the Quality of Arguments in the Sense of Justification
At first glance, determining the quality of arguments that consist only of a premise-
conclusion sequence so that one or more premises support or justify a conclusion
should be easy. There are just two things that can be wrong with such a simple
argument: either the reasons (premises) provided are not true (or not acceptable), or
the reasons do not succeed in justifying the conclusion. Both criteria need to be
17
Carpenter and Kennedy (1988, p. 29). Since Toulmin proposed to use jurisprudence as the paradigm
case for the “use of argument,” argumentation theory seem to be preoccupied with the notion that winning
an argument or persuading an opponent is the main purpose of constructing arguments. It is time, I think,
to give up these limitations. See also Zarefsky (2016).
18
As suggested, for example, by Richard Feldman in his discussion of “Good arguments”: “If the
purpose is persuasion, then good arguments are the ones that persuade. Period” (Feldman 1994, p. 175).
19
Systematic classifications of argument purposes have been suggested by Pinto (2010), Mohammed
(2016), and Hoffmann 2016a.
123
fulfilled if the purpose of an argument is to justify a claim (Feldman 1994, p. 176;

Rosenberg 1996).
These two criteria correspond to the widely known ARS criteria which Ralph
Johnson and Anthony Blair formulated early in the history of informal logic or
argumentation theory: Good arguments can be characterized by Acceptability,
Relevance, and Sufficiency.20 These three criteria can be reduced to the two
mentioned above if we interpret Acceptability as referring to the reasons of an
argument in themselves and Relevance and Sufficiency as two criteria to assess the
justificatory relationship between reason and conclusion.21
However, things are not that simple. On the one hand, we need additional quality
criteria for arguments in the sense of justification that are more complex than just a
connection between reason and claim. On the other, there are—as I will show—
fundamental epistemological problems that are unavoidable when we try to apply
these general considerations in the evaluation of concrete arguments.
4.1 Quality Standards for More Complex Justifications
Increased complexity of arguments in the sense of justification can result from the
complexity of an argument’s conclusion. Here is an example. The following
statement is a proposal that a team of students developed over the course of a
semester as a possible solution of the “Tri-state water war” in which Alabama,
Georgia, and Florida are involved for more than two decades:
All water users and beneficiaries in the Apalachicola-Chattahoochee-Flint
(ACF) River Basin should pay a water tax into a Tri-State Water
Compensation Fund according to the size of their stake in the water system.
In the beginning 80% of the Fund should be used for water conservation and
water storage projects (technology and infrastructure) prioritized according to
their efficiency and 20% should be used for building up an emergency fund for
compensation payments for future sectoral and environmental damage caused
by low water flow. The Fund’s ratio of funding for water conservation/water
storage projects and compensation savings should be adjusted every 4 years
based on an assessment of actual and expected low-water-flow damages over a
8-year period and of proposed conservation and storage projects. Water
withdrawal should additionally be taxed with a flat fee when water levels are
20
Johnson and Blair (1977); see also Govier (2010, p. 87). The authors suggested a few modifications
regarding these ARS criteria in the Preface of the book’s second edition: Johnson and Blair 2006 [1977],
pp. xiii–xv.
21
See also Freeman (2005, p. 19). It has to be noted that my understanding of “acceptability” differs
from Johnson and Blair’s as developed in the first edition of Logical Self-defence. Since they discuss
“begging the question” here as violating the principle of acceptance (Johnson and Blair 2006 [1977],
p. 65), it is clear that for them “accepting” includes “acceptable as a justification.” I prefer instead—
following Feldman (1994, p. 176), Rosenberg (1996), and Govier (2010, p. 87)—to interpret “acceptable”
solely in the sense that Johnson and Blair themselves characterize in the book’s second edition as follows:
“One way to express this point is to say that in some contexts an argument’s premises are worthy of
acceptance only if they are known or reasonably believed to be true by the arguer, and can be shown to
the audience to be true or reasonable to believe. One way to show that an argument’s premise is
unacceptable is to show that it is false” (Johnson and Blair 2006 [1977], p. xiii).
123
high and with an exponentially increasing fee that kicks in when a “buffer
zone” of water flow is reached (which should be defined, based a
comprehensive ACF basin water system model, as the area between two
water levels: the seasonally changing not-damage-causing water flow level
and a significant percentage above it). Damages caused by low water flows
and lake levels should be compensated by the Fund to actors in the ACF basin
in the order of their position in a ranking system that is based on their lack of
opportunity to control their water supply.
Wicked problems and conflicts in which multiple parties negotiate often require
proposals of such complexity. Even though all the components of such a proposal
can be justified by their own argument, it is clear that the justification of the
proposal as a whole requires to see these components as mutually dependent: The
proposed solution works only as a package because everything is connected by one
underlying mechanism of payments into and out of a shared water fund.
The complexity of an argument’s conclusion leads to two problems that are
relevant for assessing its quality. First, it may happen that such a conclusion entails
logical contradictions or conflicting components. If that is the case, then a
justification is either impossible—in case of a logical contradiction—or at least
insufficient. Secondly, conclusions that consist of multiple elements can only be
justified by a more or less complex network of arguments. Not only needs each
component of such a conclusion to be justified, but it is also necessary to get the
overall structure right. This includes organizing large amounts of information and
principles (e.g., utilitarian benefit maximization or basic rights) so that chains of
arguments—reasons justified by arguments whose reasons are justified by further
arguments, and so on—are most clearly and convincingly structured; but it also
includes realizing the correct choice of structural options such as knowing when a
proposition can only be justified by a certain combination of reasons, as in the
example depicted in Fig. 1.
Cost–benefit arguments like the one in Fig. 1 need at least two reasons: one
outlining the costs, the other the benefits. Any argument whose conclusion puts a
certain number of different things into a relation requires minimally the same
number of “linked” reasons, whereby “linked” means that all these reasons need to
be true or acceptable to justify the conclusion.
It seems to me that it should be possible—at least in principle—to determine
additional quality criteria besides the ARS criteria that would allow the evaluation
of more complex arguments. At least with regard to structural options a lot of work
has already been done (Walton et al. 2008; Freeman 2011), even though many
points are still controversial (Lumer 2016; Hitchcock 2015). Anyway, I do not want
to go into more detail here regarding complexity issues because more important
seems to me that there are still substantial difficulties with regard to assessing the
quality of simple arguments in the sense of justification. In the following, I will first
focus on the acceptability of reasons, then on the justificatory relation between
reasons and conclusion, and finally on problems related to identifying arguments in
“the real world.”
123
Fig. 1 The main argument of the Stern Report on the economics of climate change (Stern 2007). Quotes
and data from http://webarchive.nationalarchives.gov.uk/20100407172811/http://www.hm-treasury.gov.
uk/d/Summary_of_Conclusions.pdf. The argument has been reconstructed using the AGORA software
which constructs arguments in logically valid form, in contrast to the Report. The purpose of the logical
reconstruction is to stimulate reflection on the question whether the reasons are sufficient to justify the
conclusion by representing the justificatory relation itself in form of a premise (Hoffmann 2016b). It
should be noted that the conclusion should better say that the “net-benefit” of acting outweighs the net-
benefit of not acting
4.2 When are Reasons Acceptable?
When Johnson and Blair introduced the ARS criteria in Logical Self-Defense, they
wrote that the “acceptability” of a premise “concerns the relationship of the premise
to the audience” whereas its “truth” concerns the “relationship of the premise to the
world” (Johnson and Blair 2006 [1977], p. 76). It is well known that using
“acceptance by an audience” as a quality criterion for good arguments is
problematic because the audience might be especially stubborn or gullible.
“Stubborn people can fail to be persuaded by good arguments. Gullible people
can be persuaded by bad arguments” (Feldman 1994, p. 168). In order to avoid the
kind of relativism that is inevitable when quality depends on arbitrary characteristics
of an audience, acceptability has usually been heavily qualified. It has been
suggested that the audience has to be conceptualized as a “universal audience”
(Perelman and Olbrechts-Tyteca 1969 [1958]) or as “reasonable” (van Eemeren and
Grootendorst 2004, p. 1; Blair and Johnson 1987, p. 50), or that acceptability has to
be specified as “rationally acceptable” in the sense of being “reasonable for the
person to whom the argument is addressed to accept” certain premises (Govier
2010, p. 116). Nevertheless, as Richard Feldman showed in detail, these attempts to
rescue the notion of audience acceptability run into serious problems: “different
members of an audience may have different bodies of evidence and thus have
different propositions justified to them” (Feldman 1994, p. 171); “many arguments
don’t have any definitive intended audience” (p. 172); and many premises require
specific knowledge which the evaluator might not have (pp. 174–175).
123
Based on the last point, Trudy Govier concludes that “general standards that will
give complete and detailed guidelines for determining the rational acceptance of
premises” are “not possible” (Govier 2010, p. 116). What is possible, for her, is to
support premises by further arguments, by a priori and common knowledge, or by
testimony or authority. There are certainly premises that are—at least at a certain
point in time—generally acceptable, but equally likely is the assumption that many
which we may find at the end of chains of arguments are not. There will always be
premises that remain contested by reasonable people so that the same argument is
valued as good by one party but as bad by the other.
Truth, on the other hand, or the “relationship of the premise to the world”
(Johnson and Blair 2006 [1977], p. 76) is no less problematic. As Hamblin already
pointed out, it “is not enough for the premisses of an argument to be true: they must
also be known to be true” (Hamblin 1970, p. 236). But real human beings are not
omniscient, they do not have perfect knowledge. Thus, a distinction between
“different knowing subjects” needs to be made, at least when arguments are used
across different epistemic contexts as it is often the case when one person tries to
convince another one (p. 239). Even though there are many statements that can be
claimed to be true without any problem—it is uncontroversial that this sentence
contains one word that starts with a “u”—there are many others whose truth is
contested. Since our knowledge as evaluators of not further justified premises is
always limited, we can never exclude the possibility that a premise we consider to
be true or false might be contested by other people with good reasons.
A convincing way out of this dilemma—at least up to a certain point—has been
developed by James Freeman in Acceptable Premises (Freeman 2005). He defines
acceptability in terms of “presumption”: A premise is rationally acceptable at a
certain time if a person that is committed to challenge this premise presumes that it
is acceptable (p. 32). There is such a
presumption in favor of a statement S at a point p in a dialectical exchange for
the proponent P of that exchange if and only if there is a point p’ of the
exchange accessible from p for P and P has answered at p’ all challenges
raised at p against S. (p. 30; his emphases)
For Freeman, this criterion of premise acceptability in terms of presuming that it is
true if no reasonable challenges of its truth are conceivable does not entail
relativism (pp. 27, 31–32). This, however, is a question whose answer depends on
how we define relativism. It is certainly the case that the application of this criterion
ensures that the evaluator of a premise does her best do disprove it before assuming
it to be true. But human evaluators are not omniscient. They can never know
whether there might be a dialectical exchange with a challenger who can indeed
disprove a premise. Thus, there is no way to avoid the conclusion that also
acceptability as presumption depends on how much a particular argument evaluator
knows at a certain point in time.
At this point I would follow Richard Feldman who argued that while certain
forms of relativism can be avoided, the dependence on particular epistemic
situations cannot. “The relativism here is simply a consequence of the fact that
different people can be in different epistemic situations. It does not imply that any
123
belief is justified if you think it is” (Feldman 1994, p. 182). For Feldman, there is
nothing objectionable in this form of relativism (pp. 181–183). But it has to be clear
that the unavoidable relativism brought in by varying epistemic situations limits the
possibility of argument assessment. There cannot be a final or objective answer to
the question when a premise is acceptable and when it is not as long as an argument
is good only for a specific person, as in Feldman’s epistemic account of good
arguments (p. 179). Feldman assumes that this account explains “the core sense of
the phrase ‘good argument’ as it is used in standard contexts” (p. 187). But that is
questionable. I agree that this is the only sense that is defensible, but what people
expect when they hear the phrase “good argument” is an empirical question, not a
philosophical one.
4.3 Assessing the Justificatory Relation Between Reasons and Conclusion
A fundamental problem that any argument assessment has to address is the fact that
reason and conclusion in an argument talk about different things, at least as long as
the reasons are relevant for the conclusion. In case the reason says exactly the same
as the conclusion, or the argument is begging the question in the sense that the
reason is “equivalent in meaning to the conclusion” (Johnson and Blair 2006 [1977],
p. 84), the reason cannot be relevant for the conclusion. We can follow here Johnson
and Blair and define the “relevance” of a reason or set of reasons for the conclusion
as “having bearing on the truth of the claim at issue” (Johnson and Blair 2006
[1977], p. xiv).
In which way a relevant reason and a conclusion claim something different can
be seen in the original cost–benefit argument from the Stern Review on the
Economics of Climate Change that has been reconstructed in logical form in Fig. 1.
To put it more simple than quoted above, the conclusions says: The benefits of
actions against climate change outweigh their costs. And the two reasons claim:
non-acting would cost 5% of global GDP whereas the actions would cost 1% of
global GDP. Thus, what is claimed on the side of the reasons is different from what
is claimed in the conclusion. What we find in between these two sides is what I call
the “justificatory relation.”
In the literature, the quality of justificatory relations has been addressed by two
approaches. The first one, going back to Aristotle’s syllogisms, focuses on certain
structures of arguments that are defined so that the structure alone guarantees the
quality of the justificatory relation. The second approach looks at the content of
what is claimed in reason and conclusion and tries to determine—as Johnson and
Blair define their criterion of sufficiency—whether the premises of an argument
supply “all the grounds that are needed to make it reasonable to believe its
conclusion” (Johnson and Blair 2006 [1977], p. xv).
Richard Feldman takes the first route in his “epistemic account of good
arguments.” Combining the problem of premise acceptability and criteria for an
acceptable justificatory relation, he proposes the following evaluation criterion:
123
An argument is a good argument for person S if (i) S is justified in believing

the conjunction of all the premises of the argument, (ii) S is justified in
believing that the premises are “properly connected” to the conclusion, and
(iii) the argument is not defeated for S. (Feldman 1994, p. 179)
Feldman, in this context, does not provide more detail about what “properly
connected” can mean besides hinting at the problem of enthymemes and noting:
“arguments with missing premises are typically no good” or, alternatively, “they are
incomplete versions of good arguments” (p. 179).
It is well known that every argument can be transformed into a logically valid
argument by adding an appropriate premise. Thus, an argument of the form “A,
therefore B” can be turned into a modus ponens simply by adding to the given
premise “A” another one saying “If A, then B.” Such a transformation led to the
modus ponens in Fig. 1. But it should be clear that the transformation of the original
argument of the Stern Report into a logically valid form by adding the premise in
the middle does not improve the quality of the argument. The only thing that the
additional premise achieves is a transformation of the assessment problem. Whereas
in the original argument the problem is the justificatory relation between reasons
and conclusion, this problem has simply been transformed into a problem of
premise acceptability. The additional premise in this logical argument does nothing
more than representing this relation in this premise so that the problem is no longer
the quality of the justificatory relation but the acceptability of the premise that
represents this relation. It is true that the structure of logical arguments lets the
quality problem of justificatory relations disappear, but only to pop up again as an
acceptability problem.
This means that the logical structure alone cannot be sufficient to guarantee the
quality of an argument. This can be seen with the following argument that I found
on the public “argument analysis platform” argüman: “The death penalty is wrong”
because “The killing of a person is wrong and helps nobody.”22 Transformed into a
modus ponens, it reads:
(P1) The killing of a person is wrong and helps nobody
(P2) If the the killing of a person is wrong and helps nobody, then the death
penalty is wrong
(C) Therefore, the death penalty is wrong
Clearly, (P2) is not acceptable because a procedure that is performed according to
the rule of law—whatever we think about such a law—does not fall under “killing”
as the term is defined in legal contexts. But this fault was already in the original
argument. As we see, it cannot be corrected by using a logically valid structure. The
logical form is not sufficient to make a bad argument better.
As Ralph Johnson points out in a discussion of enthymemes, identifying missing
premises is much easier when there is a standard of completeness such as logical
validity “where the notion of structure is tightly circumscribed” than with regard to
22
http://en.arguman.org/the-death-penalty-is-wrong, accessed Sep 9, 2017.
123
the kind of arguments informal logic is interested in. He sees informal logic in a
“state of indeterminacy … with respect to the issue of structure” (Johnson 2000,
pp. 132–133). Even though a lot of work has been done in the meantime on the
structures of non-logical arguments that can be used for identifying missing
premises (Walton et al. 2008; Freeman 2011), a crucial question is whether getting
the structure of an argument right can ever be sufficient to achieve high quality
regarding its justificatory relation.
Let’s have a look at argumentation schemes as one way to structure arguments in
non-logical form. Leaving aside the dialectical context in which Walton, Reed, and
Macagno developed their argumentation schemes (Walton et al. 2008), we could
transform each of these schemes into a rich argument structure by translating the
critical questions they determined for each scheme into additional premises. For
example: Walton et al. (2008) determine the scheme for “argument from expert
opinion” as follows:
Major Premise: Source E is an expert in subject domain S containing
proposition A.
Minor Premise: E asserts that proposition A is true (false).
Conclusion: A is true (false). (Walton et al. 2008), p. 310)
If we would add to this scheme additional premises—based on the authors’ critical
questions—a richer structure that could then be used as a standard for quality
assessment might list the following premises:
● Source E is an expert in subject domain S containing proposition A

● E asserts that proposition A is true (false)
● E is credible as an expert source
● E is an expert in the field that A is in
● E asserts something that implies A
● E is personally reliable as a source
● A is consistent with what other experts assert
● E’s assertion is based on evidence X.
Compared to the original argumentation scheme, this enriched structure would

certainly reduce the gap between the original two premises and the conclusion. And
knowing about the possibility of these additional premises would certainly help to
construct better arguments, just as the dialectical device of critical questions does.
However, when it comes to using this enriched structure to assess the quality of a
given argument, these additional premises can only be taken into account in case we
actually know that they are true. But this is a question of the argument’s content, not
of its structure. This, again, means that structure alone is not sufficient to guarantee
an argument’s quality.
Although realizing a certain structure is, thus, not sufficient for a “proper
connection” (Feldman 1994, p. 179), one might argue that it is at least a necessary
condition for a good argument. But this claim can be refuted by the fact that we can
have perfectly good arguments that do not realize any particular structure besides
123
clearly distinguishing reason and conclusion. This can be seen in another argument,
again from the argüman platform, in which the conclusion “The death penalty is
wrong” is justified by the reason: “it’s something you cannot undo if [the person
punished is] later proven innocent.”23 This argument seems good without being
structured according to either a logical form or a non-logical argumentation scheme.
Thus, having an appropriate structure is neither sufficient nor necessary for good
arguments.
These arguments against the first approach to assess the quality of justificatory
relations by means of particular structures lead immediately to the second, to look at
the content of an argument. However, following this approach means that we face
again the same epistemological problems that we discussed with regard to premise
acceptability. The same argument can be assessed as either good or bad, depending
on the epistemic situation of the evaluator. One person might think that the premise
in a logical argument that represents its justificatory relation is acceptable, another
might disagree; or one might think a certain critical question can be answered
affirmatively, another might object.
Related to this epistemological problem is another one which focuses specifically
on bridging the “gap” between reasons and conclusion. Making the leap between
reasons and conclusion requires knowledge, and the ability to make this leap
depends on the degree of knowledge we have available in a particular situation. This
point can be supported by an example regarding the evidence-hypothesis
relationship that Helen Longino provided in Science as social knowledge:
For instance, a set of footprints preserved in volcanic ash at Laetoli and
discovered by Mary Leakey in 1976 is cited as evidence that bipedal hominids
had developed at least 3.59 million years ago. Here, too, assumptions are
required to connect the data to the hypothesis. In this case the assumptions are
generalizations embedded in coherent and accepted understandings (theories)
of sets of phenomena. For instance, simple and readily made observations
enable us to gauge the pressure exerted by the feet of fully upright walkers and
quadrupeds or knucklewalkers, and to establish the foot design necessary for
these forms of locomotion. This facilitates the inference that the prints were
left by hominids rather than by incompletely bipedal or nonbipedal creatures.
Contemporary physics and chemistry have seen the development of a number
of different but mutually consistent tests for dating fossils and other remains.
This mutual coherence supports reliance on the potassium-argon test that
assign to the volcanic tuff an age of 3.59 to 3.77 million years. (Longino 1990,
p. 110)
4.4 Where is the Argument in all this?
A third problem besides premise acceptability and the quality of justificatory

relations is that the quality of an argument can only be assessed after two questions
are answered: Is there an argument at all? And: What exactly is it? The texts and
23
http://en.arguman.org/the-death-penalty-is-wrong; my addition in brackets.
123
speech acts (or other expressions, as discussed in the literature on visual arguments)
that we find in real life are usually far less clear than what we can see in Fig. 1 or in
any other argument that I used here as an example. As Ralph Johnson puts it: “The
turn to actual arguments has forced theorists to face up to the ill-organized,
incompletely stated, wandering-off-topic arguments often found in written and
spoken texts” (Johnson 2000, p. 128). Tens of thousands of years at campfires
produced some amazing story tellers, but that does not mean that “real people” are
well equipped to construct arguments, independently from the question whether
they are good or bad. And even if we think there is an argument, it is often not easy
to identify a clear conclusion, claim, or hypothesis in all this rambling. At other
times it is unclear what is supposed to support what. These problems are well-
known at least since Frans van Eemeren, Rob Grootendorst, Sally Jackson, and
Scott Jacobs published Reconstructing argumentative discourse (van Eemeren et al.
1993).
In situations like these we could apply the principle of charity, that is, interpret a
given expression generously. Trudy Govier distinguished two versions of the
principle, a “very generous” one and a “principle of Modest Charity” (Govier 2010,
p. 52):
On a very generous principle of charity, not supported here, we would make
out an argument to be as reasonable and plausible as we could, always giving
the arguer the benefit of the doubt. On a more modest principle of charity,
recommended in this text, we would avoid attributing to an arguer loose
reasoning and implausible claims unless there is good evidence, in the
presented speech or writing, for doing so. (Govier 2010, p. 55)
The risks of being too generous, as she points out, include that we may read “too
many ideas of our own” into someone’s expressions, so the principle of charity
should be balanced with the principle of accuracy (pp. 51, 55). The question is,
however, whether we should use even the modest version when it comes to
assessing the quality of arguments. For Johnson the answer to this question depends
on what we are after. If the purpose of an analysis is “to get at the truth of the
matter,” that is, to develop the best possible argument for an issue, then “the
argument should be reconstructed in a way that maximizes this goal, regardless of
whether the argument is the one intended by the arguer” (Johnson 2000, p. 132). But
if “the purpose of analyzing the argument is to criticize the argument so that the
arguer may improve it,” then—I would say—not even the modest principle of
charity should be applied. The point of assessing arguments is to measure by how
much they deviate from a certain standard of good argument; the point is not to
make a given argument better than it actually is.
But what shall we do if we are not even sure whether there is an argument at all
in an ill-organized and ambivalent speech act? The problem is: When we are
looking for an argument, we inevitably apply certain search criteria that are
determined by our understanding of what an argument is: a certain structure, certain
indicator words like “because” and “therefore,” and so forth. If our goal is to
interpret a given speech act as an argument, we are primed to find what we are
looking for. The cognitive problem is that we can never know to what degree we
123
apply the principle of charity unconsciously. Of course, it is also possible that we

prematurely reject something because we are not able or willing to make sense of it,
or that we do not interpret an expression as an argument that was intended as one.
There are two problems in all these situations. First, we do not know to what
degree our interpretations are determined by what we bring to the situation. Even if
we try our best and engage in a thorough and fair analysis, it is questionable whether
we are able to control our interpretations completely. Second, if we interpret an
expression as an argument in the sense of justification, we need to interpret it in a
way that it fits minimally into the structure reason—justificatory relation—
conclusion. This means, however, that the assessment contains at least two
operations: translating the expression into this structure and then assessing the
argumentative strength of the argument as it has been structured. The problem is
that we cannot exclude the possibility that both operations influence each other; we
might reconstruct the argument in a way that makes it easier to find either a good or
a bad argument in it. All this means, again, that relativism seems to be as
unavoidable here as we found it in the assessment of premise acceptability and of
the justificatory relation.
5 Conclusion
How to assess the quality of arguments is a problem at least since Aristotle wrote
about it, and many more scholars have written about it since then. With regard to
argument in the sense of justification, the main result of the present critical
discussion about argument assessment is that there are unavoidable epistemic
limitations. The fundamental problem that any evaluator of arguments will
encounter is limited knowledge. Whether we try to determine the acceptability of
premises, or answer the question if given reasons are sufficient to justify a
conclusion, or if there is an argument at all in some ill-organized expressions, and if
so, what exactly it is, in all these situations our judgements and interpretations are
often limited by the epistemic situation in which we find ourselves at a certain point
in time. Usually we do not know everything that would be required to evaluate an
argument’s quality objectively.
This point has already been made by Richard Feldman in 1994 so that—with
regard to argument as justification—this contribution mainly updates Feldman’s
arguments and provides some refinements, for example regarding his not quite
sufficient considerations about the appropriate structure of arguments. The main
points with which, I think, the present contribution goes beyond this extensive
literature can be summarized as follows:
1. Since “arguments” are different things for different traditions—I proposed to

distinguish argument as controversy, as debate, and as justification—it is
necessary to develop quality standards for each of them.
2. Even within the same conceptualization of argument, different purposes for
which arguments are used require different standards of good arguments. I
showed this with regard to different sets of rules that have been developed, in
123
the context of argument as debate, for parliamentary deliberation and for the
platform ChangeMyView.
3. A new field of study is indicated by my considerations about the complexity of
conclusions and the corresponding need for quality criteria that have nothing to
do with reasons or the justificatory relation between reason and conclusion.
Conclusions themselves can be so badly formulated that the question whether
reasons are relevant or sufficient cannot be answered.
4. Finally, the question discussed in the last section—whether it is possible to
control unconscious applications of the principle of charity—suggests that there
might be more aspects of argument assessment that should be studied from a
cognitive science point of view.
The past few years saw important publications in this last area, from the
“argumentative theory of reasoning” (Mercier and Sperber 2017) to The Knowledge
Illusion (Sloman and Fernbach 2017). Both approaches emphasize the social
character of arguing. What this empirical research on cognitive limitations of real
people means for the possibility of argument assessment remains to be seen. Future
research should focus in particular on the question whether it is possible to separate
good from bad arguments in highly polarizing conflicts. Given research on so-called
motivated reasoning, a well-studied cognitive bias that refers “to the tendency of
people to conform assessments of information to some goal or end extrinsic to
accuracy” (Kahan 2013, p. 408), the notion of “objective” argument quality might
be even more elusive than discussed in this contribution.
Acknowledgements This research has been supported by a grant from the National Science Foundation
(Cyberlearning and Future Learning Technologies, Award 1623419). I am thankful for important
feedback that Bryan Norton, Justin Biddle, Matt Cox, and two anonymous reviewers provided to earlier
versions of this paper.
References
Aakhus, M., and M. Lewiński. 2017. Advancing polylogical analysis of large-scale argumentation:
Disagreement management in the fracking controversy. Argumentation 31 (1): 179–207. https://doi.
org/10.1007/s10503-016-9403-9.
Andriessen, J.E.B., M. Baker, and D.D. Suthers (eds.). 2003. Arguing to learn. Confronting cognitions in
computer-supported collaborative learning environments. Dordrecht: Kluwer Academic Publishers.
Aristotle. (An. Pr.) 1996 [1938]. Analytica priora. In Aristotle. The categories. On interpretation. Prior
analytics, ed. H. Tredennick, 24a10–70b. Cambridge: Harvard University Press, Heinemann.
Aristotle. (Soph. el. [Forster]) 1955. Sophistici elenchi On sophistical Refutations. On coming to be and
Passing-Away. On The Cosmos. Cambridge: Harvard University Press.
Bächtiger, A. 2005. The real world of deliberation: A comparative study of its favorable conditions in
legislatures. Bern: Haupt.
Barth, E.M., and E.C.W. Krabbe. 1982. From axiom to dialogue. Berlin: Walter de Gruyter.
Benn, N., and A. Macintosh. 2012. Making sense of macro-and micro-argumentation in policy-
deliberation: Visualisation techniques and representation formats. In Computational models of
argument: proceedings of comma 2012, eds. B. Verheij, S. Szeider, S. Woltran, 71–82.
Bentahar, J., B. Moulin, and M. Belanger. 2010. A taxonomy of argumentation models used for
knowledge representation. Artificial Intelligence Review 33 (3): 211–259. https://doi.org/10.1007/
s10462-010-9154-1.
123
Besnard, P., A. Garcia, A. Hunter, S. Modgil, H. Prakken, G. Simari, and F. Toni. 2014. Introduction to
structured argumentation. Argument & Computation 5 (1): 1–4. https://doi.org/10.1080/19462166.
2013.869764.
Blair, J.A., and R.H. Johnson. 1987. Argumentation as dialectical. Argumentation 1 (1): 41–56.
Bochenski, J.M. 1970. A history of formal logic, 2nd ed. New York: Chelsea Pub. Co.
Buckingham Shum, S. 2003. The roots of computer-supported argument visualization. In Visualizing
argumentation: Software tools for collaborative and educational sense-making, ed. P.A. Kirschner,
S.J. Buckingham Shum, and C.S. Carr, 3–24. London: Springer.
Caluwaerts, D., and K. Deschouwer. 2014. Building bridges across political divides: Experiments on
deliberative democracy in deeply divided Belgium. European Political Science Review 6 (3): 427–
450. https://doi.org/10.1017/s1755773913000179.
Carpenter, S.L., and W.J.D. Kennedy. 1988. Managing public disputes: A practical guide to handling
conflict and reaching agreements, 1st ed. San Francisco: Jossey-Bass.
Conklin, J. 2006. Dialogue mapping: Building shared understanding of wicked problems. Hoboken, NJ:
Wiley.
Dung, P.M. 1995. On the acceptability of arguments and its fundamental role in nonmonotonic reasoning,
logic programming and n-person games. Artificial Intelligence 77 (2): 321–357.
van Eemeren, F.H., R. Grootendorst, S. Jackson, and S. Jacobs. 1993. Reconstructing argumentative
discourse. Tuscaloosa: University of Alabama Press.
Feldman, R. 1994. Good arguments. In Socializing epistemology. The social dimensions of knowledge, ed.
F.F. Schmitt, 159–188. Lanham, MD: Rowman & Littlefield Publishers.
Fisher, A. 2001. Critical thinking: An introduction. Cambridge: Cambridge University Press.
Freeman, J.B. 2005. Acceptable premises. An epistemic approach to an informal logic problem.
Cambridge: Cambridge University Press.
Freeman, J.B. 2011. Argument structure: Representation and theory. Dordrecht: Springer.
Goddu, G.C. 2009. Refining Hitchcock’s definition of ‘argument’. In Proceedings of the 8th international
conference of the Ontario society for the study of argumentation (OSSA), 1–12. Retrieved from http://
scholar.uwindsor.ca/ossaarchive/OSSA8/papersandcommentaries/55/.
Govier, T. 2010. A practical study of argument, 7th ed. Belmont, CA: Cengage Learning.
Groarke, L., C.W. Tindale, and J.F. Little. 2008. Good reasoning matters!: A constructive approach to
critical thinking, 4th ed. New York: Oxford University Press.
Hamblin, C.L. 1970. Fallacies. London: Methuen.
Hansen, H.V. 2015. Fallacies. The Stanford encyclopedia of philosophy (summer 2015 edition). Retrieved
5 Feb 2017, from https://plato.stanford.edu/archives/sum2015/entries/fallacies/.
Hitchcock, D. 2007. Informal logic and the concept of argument. In Philosophy of Logic, ed. D. Jaquette,
101–129. Amsterdam: Elsevier.
Hitchcock, D. 2015. Freeman’s syntactic criterion for linkage. Informal Logic 35: 1–31. Retrieved from
http://ojs.uwindsor.ca/ojs/leddy/index.php/informal_logic/article/view/4234.
Hoffmann, M.H.G. 2016a. Reflective argumentation: A cognitive function of arguing. Argumentation 30
(4): 365–397. https://doi.org/10.1007/s10503-015-9388-9.
Hoffmann, M.H.G. 2016b. Stimulating reflection and self-correcting reasoning through argument
mapping: Three approaches. Topoi. An International Review of Philosophy. https://doi.org/10.1007/
s11245-016-9408-x
Jackson, S. 2015. Design thinking in argumentation theory and practice. Argumentation 29 (3): 243–263.
https://doi.org/10.1007/s10503-015-9353-7.
Johnson, R.H. 2000. Manifest rationality: A pragmatic theory of argument. Mahwah, NJ: Lawrence
Erlbaum Associates.
Johnson, R.H. 2009a. The implications of the dialectical tier for critical thinking. In Critical thinking
education and assessment: Can higher order thinking be tested?, eds. L. Groarke, R.H. Johnson, J.
Sobocan, and F.S. Ellett, 55–74. London: Althouse Press.
Johnson, S.L. 2009b. Winning debates: A guide to debating in the style of the world universities debating
championships. New York: International Debate Education Association.
Johnson, R.H., and J.A. Blair. 1977. Logical self-defense: the craft of evaluating persuasion in everyday
life. Toronto, ON: McGraw-Hill Ryerson.
Johnson, R.H., and J.A. Blair. 2006 [1977]. Logical self-defense. New York: International Debate
Education Association.
Kahan, D.M. 2013. Ideology, motivated reasoning, and cognitive reflection. Judgment and Decision
Making 8 (4): 407–424.
123
Kirschner, P.A., S.J. Buckingham Shum, and C.S. Carr (eds.). 2003. Visualizing argumentation: Software
tools for collaborative and educational sense-making. London: Springer.
Klein, M., P. Spada, and R. Calabretta. 2012. Enabling deliberations in a political party using large-scale
argumentation: A preliminary report. In Proceedings of the 10th international conference on the
design of cooperative systems. Retrieved from https://www.researchgate.net/publication/263307756.
Kuhn, D. 1991. The skills of argument. Cambridge: Cambridge University Press.
Lewiński, M. 2015. Argumentative discussion: The rationality of what? Topoi. https://doi.org/10.1007/
s11245-015-9361-0.
Lewinski, M., and M. Aakhus. 2014. Argumentative polylogues in a dialectical framework: A
methodological inquiry. Argumentation 28 (2): 161–185. https://doi.org/10.1007/s10503-013-9307-
x.
Longino, H.E. 1990. Science as social knowledge. Values and objectivity in scientific inquiry. Princeton,
NJ: Princeton University Press.
Lumer, C. 2016. Walton’s argumentation schemes. Paper presented at the Argumentation, Objectivity and
Bias. In Proceedings of the 11th international conference of the Ontario society for the study of
argumentation (OSSA), May 18–21, 2016, Windsor, Canada.
Mercier, H., and D. Sperber. 2017. The enigma of reason. Cambridge, MA: Harvard University Press.
Mohammed, D. 2016. Goals in argumentation: A proposal for the analysis and evaluation of public
political arguments. Argumentation 30 (3): 221–245. https://doi.org/10.1007/s10503-015-9370-6.
Perelman, C., and L. Olbrechts-Tyteca. 1969 [1958]. The new rhetoric. A treatise on argumentation (trans:
Wilkinson, J., Weaver, P.). Notre Dame, IN: University of Notre Dame Press.
Pinto, R.C. 2010. The uses of argument in communicative contexts. Argumentation 24 (2): 227–252.
https://doi.org/10.1007/s10503-009-9174-7.
Plato. (Phd.) 1900 ff. Phaedo. In Platonis opera, ed. I. Burnet, Vol. I, 57–118. Oxford: E Typographeo
Clarendoniano.
Plato. (Tht.) 1900 ff. Theaetetus. In Platonis opera, ed. I. Burnet, Vol. I, 142–210. Oxford: E
Typographeo Clarendoniano.
Quintilian. 1921. The Institutio oratoria of Quintilian, with an English translation by H. E. Butler (trans:
Butler, H.E.). Cambridge, MA: Harvard University Press.
Rittel, H.W.J., and D. Noble. 1989. Issue-based information systems for design. University of California
at Berkeley working paper, 492.
Rittel, H.W.J., and M.M. Webber. 1973. Dilemmas in a general theory of planning. Policy Sciences 4:
155–169.
Robert, H.M. 1915. Robert’s rules of order revised for deliberative assemblies. Chicago: Scott.
Robert, H.M., and S.C. Robert. 2011. Robert’s rules of order newly revised, 11th ed. Philadelphia, PA: Da
Capo Press.
Rosenberg, J.F. 1996. The practice of philosophy. A handbook for beginners, 3rd ed. Upper Saddle River,
NJ: Prentice Hall.
Scheuer, O., F. Loll, N. Pinkwart, and B.M. McLaren. 2010. Computer-supported argumentation: A
review of the state of the art. International Journal of Computer-Supported Collaborative Learning 5
(1): 43–102.
Schneider, J., T. Groza, and A. Passant. 2013. A review of argumentation for the Social Semantic Web.
Semantic Web 4 (2): 159–218. https://doi.org/10.3233/SW-2012-0073.
Sextus, E. 1912. Sexti Empirici Opera. Recensuit Hermannus Mutschmann. Vol. I, Pyrroneion
Ypotyposeon. libros tres continens. Lipsiae: in aedibus B. G. Teubneri.
Sloman, S.A., and P. Fernbach. 2017. The knowledge illusion. Why we never think alone. New York:
Riverhead Books.
Steiner, J. 2012. Newest version of Discourse Quality Index (DQI). In Foundations of deliberative
democracy: Empirical research and normative implications, ed. J. Steiner, 268–271. Cambridge:
Cambridge University Press.
Steiner, J., A. Bächtiger, and M. Spörndli. 2004. Deliberative politics in action: Analyzing parliamentary
discourse. Cambridge: Cambridge University Press.
Stern, N.H. 2007. The economics of climate change. The stern review. Cambridge: Cambridge University
Press.
Toulmin, S. 2003 [1958]. The uses of argument (updated ed.). Cambridge: Cambridge University Press.
van Eemeren, F.H., B. Garssen, E.C.W. Krabbe, A.F. Snoeck Henkemans, B. Verheij, and J.H.M.
Wagemans. 2014. Handbook of argumentation theory, 1st ed. New York: Springer.
123
van Eemeren, F.H., and R. Grootendorst. 2004. A systematic theory of argumentation. The pragma-
dialectical approach. Cambridge: Cambridge University Press.
Walton, D.N., C. Reed, and F. Macagno. 2008. Argumentation schemes. Cambridge: Cambridge
University Press.
Wodak, R., and V. Koller. 2010. Handbook of communication in the public sphere. Berlin: Mouton de
Gruyter.
Zarefsky, D. 2016. Is dialogue the most appropriate model for argumentation? In Argumentation and
reasoned action: Proceedings of the first European conference on argumentation, Lisbon, ed.
D. Mohammed, M. Lewiński, 9–12 June 2015, Vol. 2, 1081–1091. London: College Publications.
123

Hoffmann2018 Article TheElusiveNotionOfArgumentQual

Uploaded by

Copyright:

Available Formats

Hoffmann2018 Article TheElusiveNotionOfArgumentQual

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Hoffmann2018 Article TheElusiveNotionOfArgumentQual

Uploaded by

Copyright:

Available Formats

Argumentation (2018) 32:213–240

The Elusive Notion of “Argument Quality”

Published online: 17 November 2017

Keywords Argument assessment · Argument definitions · Argument quality ·

& Michael H. G. Hoffmann

understanding of fallacies is reflected in Hamblin’s definition when he writes in his

Since reflection on arguments and their role is important in many disciplines, it

2.1 Argument as Controversy Between Two Positions

Some of the definitions of argument that point to controversy include additionally

2.2 Argument as Debate on an Issue

1989). Rittel and Webber characterized “wicked problems,” in contrast to “tame

2.3 Argument as Justiﬁcation of, or Support for a Claim

This third definition of “argument” is widely shared in philosophy. The set of

2.4 More Deﬁntions of Argument

These three conceptualizations of argument—as justification, as controversy

● For an argument in the sense of justification it is both necessary and sufficient

In many theories these idealized or pure forms of argument are combined, as in

3 Three Forms of Argument: Three Standards of Quality?

4 Determining the Quality of Arguments in the Sense of Justiﬁcation

fulfilled if the purpose of an argument is to justify a claim (Feldman 1994, p. 176;

4.1 Quality Standards for More Complex Justiﬁcations

4.2 When are Reasons Acceptable?

4.3 Assessing the Justiﬁcatory Relation Between Reasons and Conclusion

An argument is a good argument for person S if (i) S is justified in believing

● Source E is an expert in subject domain S containing proposition A

Compared to the original argumentation scheme, this enriched structure would

4.4 Where is the Argument in all this?

A third problem besides premise acceptability and the quality of justificatory

apply the principle of charity unconsciously. Of course, it is also possible that we

1. Since “arguments” are different things for different traditions—I proposed to

You might also like