s e l r a h C y l l a i c e p s e t s o M . k r o w s i h t n o M A .GILBERT MULTI -MODAL ARGUMENTATION PHIL OF THE SOCIAL SCIENCES V OL 24 N R 2 .2
Once the door is opened to persuasion the entire gamut of human contextuality comes into
play, and social scientists, not just logicians (whether formal or informal) are required to fully
unravel dialogic argumentation . Argument must be seen as an interaction utilising far more than
traditional rational means to convince or persuade. In fact, the classical differentiation between
those two terms that raises fconvinceg to an honorific and fpersuadeg to a derogative must be
abandoned. Perelman, however, was not willing to go quite this far. His importation of the
Universal Audience, like Berkeleygs God, comes in to keep everything from going too far astray.
Even the most recent work in the rôle of goals in discourse has prompted J.P. Dillard
(1990) to query the rôle of flogicalityg in persuasive communications. \Three trained judges]
rated various descriptions of intereaction for this characterisitic which is identified by \offering
several realistic and compelling reasons] (p. 86), and \the degree to which the source makes use
of evidence and reason] (p. 85). Presuambly, the balance of the interaction is non-logical, and,
at the very least, is seen as seperabale from the remainder.
Consequently, it should be clear that the two assumptions, the first regarding linearity, and
the second regarding the marginalization of non-discursive forms as rational, are alive and well.
And yet, in looking, for example, at Dillardgs work, the question must be asked: by whom were
the judges trained? In whose sense of rationality? In whose system of logic? And, if persuaion
takes place is it because of the logic, or do we it subsequently, when we want to incorporate said
belief into our alethic system. Such questions are paramount for the social sciences insofar as
they imply methodological assumptions that are not made explicit in an environment when, more
and more, the direct linear tradition is supposedly being abandoned. I quote C.A. Willard.
My...proposal that argument be viewed as a form of social interaction has
proved remarkably uncontroversial; but my arguments that nondiscursive
symbolism is a core element of argumentationgs subject matter have provoked
wide dispute. This is an odd result,since I do not see how one can take the
argument-as-interaction notion seriously and still maintain that arguments are
exhaustively or uniquely linquistic communications. (1981 p. 191)
The social sciences concern themselves with people, and I agree with Willard that people
argue in an intricate matrix composed of numerous forms of communicative methods. It is
therefore essential that this matrix be examined and, even more, brought to bear as a tool of
analysis upon argumentative interaction. In this essay I abandon the two cited assumptions, and
suppose that modes of communicating, persuading, convincing and disputing that are wholly or
partially non-rational are equally integral to argumentation. In doing so, the area that is covered
by fargumentg must be reconsidered and re-defined. This essay undertakes the evaluation of the
results of suspending the two aforementioned assumptions, and offers a categorization of
argument that goes beyond both the verbal and the rational. Further, if, as the last forty years of
argumentation theory attest, we intend to treat argument as a human endeavour rather than a
logical exercise, we must make room therein for those practices used by actual arguers. In doing
this we must try as best we can to separate the normative from the descriptive, and remember at
all times that argumentation theorists are largely drawn from a highly rational professional group
that values linear reasoning above all other modes of persuasive communication. While this is
not to suggest that Western academics do not have emotions or intuitions, rationality and fbeing
rationalg are normally put forward as the correct approach for interpersonal communication
especially in formal or dialectical situations.
The trap that lies in wait for argumentation theory is that the rest of the world is not nearly
as linearly rational as Western academics. Even in North America there are millions of people M .A.GILBERT MULTI -MODAL ARGUMENTATION PHIL OF THE SOCIAL SCIENCES
who believe in the supernatural, the extra-sensory, and an entire cornucopia of religious,
mystical and New Age ontology. Indeed, I believeit is safe to surmise that far more people
believe in spirits, reincarnation, and the like than do not. If, therefore, we are concerned with
how people do in fact argue, with what sorts of material, evidence, modes of communication,
maneuvres, fallacies, andpersuasive devices people actually do draw upon, then we must go
beyond the linguistic and even beyond the rational. To do otherwise is to limit argument, by fiat,
to a narrow realm of the category of communications that persuade and/or convince. One might,
of course, insist that this is exactly what ought be done: strange and inappropriate modes of
reasoning or forms of argument have no place in good argument and ought not be encouraged.
This, of course, confuses the descriptive and the normative roles of argumentation theory as well
as supposing that we are perfectly clear on just what are the canons of good argument.
Regardless of onegs commitment to the \convince/persuade] duality, it is still important to
comprehend the range of argument as used within the world, if only to subsequently assess and
normatively categorize. In other words, we first require a taxonomy of actual (or used)
arguments before we can decide which are fgoodg and subsequently begin to proselytize on their
For the purposes of this discussion the term frationalg, used to mean reasoned, linear,
orderly, is overly narrow and restrictive. This is the sense intended in such admonitions as, \I'm
not going to argue with you if you can't argue rationally.] But it should be noted that this slogan
does not state that one is not arguing, rather that one is doing it, from the point of view of the
speaker, in an undesirable manner. In other words, being a bad argument entails, at the least,
falling into the category of argument. So, if one can fail to argue rationally, this presumably
means that there are non-rational arguments even though the speaker does not like them. In this
sense of the word, \rational] is often used as an honorific, and more importantly, as a way of
negating and/or trivializing modes of argument not in keeping with one arguer's precepts. This
sense requires the rational person to think in a certain, generally logical, way and adhere to
standards of evidence, deduction and reasoning established by a tradition that is heavily
scientific, rationalist, and male-dominated (vide Warren, 1988.)
At the heart of the rational outlook is the essential role of language, and, more particularly,
verbalization. Witness the following statement from van Eemeren & Grootendorst occurring on
the first page of Speech Acts In Argumentative Discourse (1983):
For the elimination of a difference of opinion it is important that the various
points of view are stated as clearly aspossible. As a rule this means that the
persons concerned in the difference of opinion will somehow have toverbalize
Is verbalization necessary to settle all differences of opinion? Is clarity of statement
always beneficial, or might fuzziness be integral and even desirable in certain sorts of arguments
or in arguments between certain sorts of people? 2 Presumably, when involved in intellectual
discourse of the sort that occurs in a university (and a large variety of other venues) such an
approach is warranted, or, at least, expected. But in at least as many other contexts devotion to
verbalized linearity might be inappropriate or even wrong. I claim that in many situations ego,
physicality and intuition play roles that are integral to the communicative and argumentative
situation, and that to slough these off as peripheral or, worse, fallacious, is both unwarranted and
n a c s t n e m u g r a y l r a e l 2 C proceed non-verbally, cf. Willard (1989, 96 ff.). The question is
whether we must be able to verbalize them on demand. h t d a e r b t n e i c i f f u s d r o f f a o t r e d r o n i M A .GILBERT MULTI -MODAL ARGUMENTATION PHIL OF THE SOCIAL SCIENCES V OL 24 N R 2 .5
various degrees of several or all four modes. Further, I will argue that to attempt to re-interpret
all these to the logical is prejudiced reductionism.
Before continuing it is important to clarify two basic terms. The first is fargumentg, the
second fmodeg For fargumentg I will use Willard's definition, most recently restated in A Theory
of Argumentation, (p. 1, 1989): "Argument is a form of interaction in which two or more people
maintain what they construe to be incompatible positions." Essentially, argument is communi-
cation when there is real or imagined disagreement. Explicating this, Willard says (p. 92) that
arguers, "use any or all of the communication vehicles available to them ... Once we have an
argument anything used to communicate within it is germane to an analysis of how the argument
proceeds and how it affects the arguers." In short, we have to worry less about the necessary and
sufficient conditions of argument and more about what people who are arguing actually do. In
using Willard's definition I am intentionally beginning from a very broad platform and focusing
on what should be done to understand and analyze argument taken in its broadest sense. There
are, of course, alternative definitions. Copigs definition is a classical understanding of an
argument as a Claim-Reason Complex (CRC): \An argument, in this sense is any group of
propositions of which one is claimed to follow from the others, which are regarded as providing
evidence for the truth of that one] (p. 7, 1961). The difference between Copigs definition and
Willardgs, is that while Copi certainly lays outwhat is an fargumentg, ithas no real relevance to
farguingg. That is, while it provides us with a basis for studying certain static characteristics
inherent in one slice of propositional entities, it has little or nothing to do with what happens
when peoplegs beliefs and/or attitudes come into conflict. And, it is this latter area that concerns
contemporary argumentation theory. It is the very breadth of Willardgs definition that permits
and encourages the exploration of the dynamics of argument.
What, now, is meant by a \mode of argumentation]? Arguments can be classified in as
many ways as there are scholars to classify them. One common way is to describe an argument
as logical or not. An argument that takes its information, e.g., warrant, backing, evidence, from
traditional rationalist sources, and which, in addition, is or can be put into traditional rationalist
form, viz., linguistic, is said to be in the logical mode, realm or form. Note that flogicalg is not
being used in the sense of deductive, but in the sense one has in mind when one says of a thought
or argument, \That's logical.] Paradigm logical arguments, many of which are not at all
deductively correct, are so-called dialectical arguments. (Cf. van Eemeren & Grootendorst,
1983, 1988, for an explication of classically pure dialectical argumentation.) Though this
definition of flogicalg is far from precise, we have no difficulty in understanding, at least in
paradigmatic situations, when an argument does or does not belong to the logical realm. A
similar sense of fbelongingg applies analogously to the other three modes. I.e., when we say of
an argument, bit of reasoning, claim, warrant or what have you that it is not logical, we have
little difficulty. All I am saying is that when we do so it is natural to place it in another category:
likely one of the remaining three.
An argument, then, may be said to be wholly or partially in a particular mode when its
claim, data, warrant and/or backing is drawn from that particular mode, or if these items are
communicated using a form of presentation from a particular mode. The more elements in a
particular argument drawn from a particular mode , the deeper entrenched in that mode the
argument is. Again, no claim at all is being made for purity of mode, the expectation being that
most arguments will have various elements from several modes. Nonetheless, by examining an
argument taken in the broad sense of the term, we can identify cases where one mode as opposed
to another seems to be predominant. Consider an example. M A .GILBERT MULTI -MODAL ARGUMENTATION PHIL OF THE SOCIAL SCIENCES V OL 24 N R2 .6
Example 1. John and Mary are having an argument about their vacation
plans. Mary is frustrated by John's repeatedly saying of her suggestions, "We
can't afford that." Finally, with some heat, she says, "It doesn't sound like we can
afford anything." John's face clouds over; he looks sad and embarrassed. He
turns away forlornly, head hanging down.
Is John offering an argument, a response to an argument, or performing any argumentative
move? I say that he is, and that it is offered in the emotional mode,and that to merely reduce it
to linguistic terms is to negate both the method and purpose (conscious or not) of the move.
Sometimes, granted, a non-verbal communication can be more or less directly translated into a
verbal parallel. A shrug, for instance, may clearly translate to an "I don't know." (Though it too
might be ambiguous.) However, in this example, much more is being communicated, and what
is being communicated is highly relevant to the argument considered as a whole. The kind of
information presented may defy direct translation, but that does not mean it is not an argu-
mentative move. Indeed, there is nothing that guarantees the transparency of linguistic utter-
ances▯we constantly misunderstand and misinterpret each other▯so why should such a demand
be made for non-linguistic expressions? This issue, the translatability of non-verbal
communications, is important. Presumably, if Johngs communication could be translated into
linguistic terms, and related to a claim or a premiss currently in or relevant to the argument, then
there would be no quarrel that the communication was indeed an argument. In other words, if
we can force the communication into a CRC format, then it is acceptable to call it an argument.
Unfortunately, translations of this nature are notoriously difficult. We cannot imagine such a
translation without carefully referring to the context of the argument and, perhaps, the personal
and social histories of the arguers. But this is exactly the point▯we understand the
communication as a part of an interactive argument, as a component argument of a larger
argumentative context. Any translation we might make for descriptive or discursive purposes
will rely on our understanding of the entire argumentative context, and not just on a simple
analysis of an individual item.
Alternatively, one might not say thatgs Johngs move is not an argumentative move, but that
it ought not be an argumentative move. (Burleson (1981) might be expected to say this.) But the
fact is that Mary must deal with John's upset,that it may well direct her next move, and John's
response does provide her with potentially valuable information about both his position and
himself. Moreover, if we say that John's show of emotion is fallacious then we recognize it as a
component of an argument: insofar as fallacies are incorrect or improper argument moves, they
are, ipso facto, argument moves.
Wanting to investigate alternate modes, does not imply that there is something wrong with
the logical mode. It is a basic, clear and valuable mode of argumentation vital to academic and
commercial enterprise. Given that most argumentation scholars are highly trained in the logical
mode and value it above all others it is hardly surprising that it is pre-eminent. Most of the
arguments one finds in the world, however, do not, in fact, follow a purely logical model, but
rather, I suggest, involve various modes at various times.
Having explicated these basic notions and laid out my basal assumption, it is necessary to
turn to the specific exemplification of the four modes of argument. To begin with an example M .A.GILBERT MULTI -MODAL ARGUMENTATION PHIL OF THE SOCIAL SCIENCES V OL 24 N R 2 .7
that is apparently in the logical mode, and, indeed, follows an identifiable logical pattern is
Example 2. Harry held a finger over his lips to signal for silence. He
pointed to the door with his revolver. "He's in there," he said to Jane.
"How can you be sure?" she queried.
"He had to take the left or right door before, and they both lead into that
"O.K., then," Jane replied, "I'm ready when you are."
The reasoning with which Harry reassures Jane is classically logical, and follows fairly
closely the pattern known as V-Elimination or Disjunctive Syllogism in a natural deduction
system. The pattern is as follows.
Example 3. A V B, A ▯ C, B ▯ C ▯ C
In , let A be, \he took the right door], B be, \he took the left door], and C be \he's in
that room]. Without too much difficulty we can see the connection between  and . This is
helpful in understanding the persuasive force of Harry's argument. Given, as we witnessed, that
Jane accepted the three premisses, she was persuaded that their man had to be in the room. That,
then, is the argument. But, in reality, a great deal more occurred in this argument than its
formalization shows. Harry's relation to Jane, his apparent knowledge of their surroundings, her
lack of objection or rejoinder, the participantsg likely fear and/or tension in being in a dangerous
situation all compose significant parts of the interaction. Still, the argument does lend itself to a
linear, rational mode of analysis.
A second, less formally exact, but still highly logical example is as follows.
Example 4. Shana: Let's go over to the Bijou and see that new film.
This argument is also straightforwardly logical. Zack inductively draws on experience to
conclude that their mutual objective, entrance to the show, would not be accomplished if Shana's
suggestion were followed. Even had he stated his argument by simply making a face and
pointing to the clock, the argument would still be in the same mode. In other words, being
verbal or non-verbal is not in itself either a necessary or sufficient determination of mode.
It is now necessary to present examples of arguments in the three alternative modes. These
examples purport to show that there are arguments where the sources of information, i.e.,
warrant and backing, and/or mode of presentation are essentially non-logical, and, at the same
time, are still clearly components of the argument. Before presenting them, however, it is
important to reiterate that no claim is being made for exclusivity. It is unlikely that any
argument is purely in one mode, and it is practically certain that any argument can be twisted out
of its natural shape and into some arbitrary mode. This said, an example of an argument from
the emotional mode follows.
Example 5. Jill: But why should I marry you, Jack?
JB:ecloyolitself. M A .GILBERT MULTI MODAL ARGUMENTATION PHIL OF THE SOCIAL SCIENCES VOL 24 N R 2 .8
Several points can be made about this example. First, some will think Jack's has a good
reason, while others will find it a not very compelling one. Needless to say, the strength of the
reason is independent of its mode. It might also be considered ethymematic: the argument is not
essentially emotional, but merely relies on suppressed or absent premisses for its logical
standing: \What Jack is really saying is that he will be a good husband, and that he is devoted to
Jill, and that ...] And, indeed, Jack may well assent to some such conjunction if presented to
him. But the fact that  can be paraphrased into a logical argument does not make it one; it is
an emotional one, its force and persuasive power come almost entirely from its emotional aspect.
To try and construe it otherwise is to force a square peg into a round hole. Jack's argument,
whether considered a good one or not, is perfectly well understood, and in order to understand it
we do not reduce it to logical terms. Note that there is here no objection to Jackgs having made
an argument▯there is a clear reason and claim, and in that sense is perfectly logical. However,
Jackgs reason is not logical, its source is an introspection of his emotional state. By being aware
of this we are in a better position to analyze and judge the argument.
Consider the next example.
Example 6. Paula is sitting in Professor Tome's office. She is pleading
for an 'A' in his logic course. "Don't you see," she explains plaintively, tears in
her eyes, "if I don't get an 'A' in your course I won't make medical school, and my
life will be ruined. I won't have anything left to live for."
Example  is an example of a primarily emotional argument. Paula's appeal is essentially
based on her desire to go to medical school and its emotional importance to her, as opposed to
her academic ability to meet the entrance requirements. The reason she provides Professor Tome
is the earne