Douglas Walton, a professor of philosophy at the University of Winnipeg
has researched extensively the concept of argumentation and its application
to real world reasoning. In 1998, he published a book called The New Dialectic:
Conversational Contexts of Argument (Toronto: University of Toronto Press,
1998). In this book he proposes a dialectical method of argument analysis.
Unlike the Toulmin and Scriven methods, Walton relies less on a structural
approach and more on an analytical framework. We have modified Waltons
approach by deleting some questions and adding others, particularly question
4 in Section III, Burden of Proof.
In Walton's view, an argument is evaluated according to whether an argument
in a given case is being used reasonably or not. As he writes: "
'used reasonably' means contributing to the goals of the type of discourse
which the argument is supposed to be a part of." Walton's approach
is to articulate the goals of the argument and identify the context in
which it appears. Some of Walton's language is technical and will be explained
as the four steps are explicated. In applying the framework it is important
to acknowledge that not all questions are relevant.
As Walton writes:
|The method is not meant to be applied in a purely mechanical sequence.
In some cases, it might be better to start at the fourth step in order
to get a sense of perspective. If we are confronted with a longer
sequence of argumentation, say, a whole article or even a book-length
argument, our initial job may be to work at the fourth step. The problem
here may be that the corpus is so large that we need to pare down
the details and try to grasp the main thrust, the most important line
of argument to concentrate on.
In using the four-step method, a critic
of argument has to use his judgment in ordering the sequence of steps
appropriate for the analysis of the particular case. Yet the four-step
method does provide a general perspective that can be used to check off
the various considerations that might be overlooked in judging how an
argument has been used in a context of dialogue.
Step One: Identifying
For the first step there are 6 questions to be asked. These are important
for determining the conclusions and meaning of the argument. This method
can actually supplement the structured methods you have already seen.
As with the Scriven method, we will use the example of the letter
written to the Toronto Star on breast self-examination.
1. What is the conclusion?
Finding and clarifying the conclusion is an important first step in all
forms of critical thinking. What is it that you are supposed to believe
or do if you are persuaded by the argument?
In this case we are asked not to accept
the conclusions of a controversial study that recommends that women not
be taught breast self examination (BSE) as a method to screen for breast
cancer. (The full study can be consulted in the Canadian
Medical Association Journal.)
What are the premises?
The author then sets out what she believes are reasons for not accepting
the conclusions of the study. We will simply enumerate them at this point.
Her premises are:
- Some experts disagree
- She has read the studies and finds that they are incomplete
- Studies were conducted in Russia, which is in a state of social chaos
and in China so the study populations are not comparable to Canada
- The criteria for teaching BSE are not included in the studies
- The harm of BSE has been overstated
- Those who have found lumps do not believe BSE is useless
- Despite the fact that most lumps that are biopsied are benign, it
is a great comfort to know this
- BSE is simple and Canadian women are capable of learning BSE
- BSE increases the chances that breast cancer is found early and at
a treatable stage
2. Are there too few premises? Check for omissions of relevant premises.
The author has provided 9 reasons to support her claim. It is generally
not clear how many premises are required in any argument. The task of
critical analysis is to determine this. It is not evident at this stage
whether there are omissions.
3. Is it a question that needs attention? What are the presuppositions?
Should the presuppositions of the question be acceptable to the respondent?
4. Is there ambiguity too many premises? Check for equivocation.
Is the conclusion unambiguously indicated?
There is no ambiguity in the sense of using words with different senses.
The conclusion is unambiguously stated in the first paragraph.
5. Is it an enthymeme? Identify missing (unexpressed) premises.
There are several enthymemes (unstated assumptions) that can be summarized
in point form:
- That all research from places with social unrest cannot be considered
- That anxiety from diagnostic and screening tests is equivalent
- That BSE is analogous to a blood test
- That those who find a lump are similar to those who examine their
breasts and miss lumps
- That finding more breast masses leads to better outcomes, which is
precisely the issue at hand.
6. What is the ultimate conclusion? Analyse the sequence or reasoning.
Are there points where relevance of a line of argument are open to question?
The arguments ultimate conclusion is that BSE should be supported
as part of breast cancer screening. The supporting reasons, in concert,
provide a compelling argument to support this as against the recommendations
of the Canadian Task Force on Preventive Health Care.
There are several places where relevance is open to question.
- It is not clear how the Harvard doctors are to be believed, or how
relevant their views are.
- It is arguable that the social chaos of Russia is relevant to the
quality of the science.
- The comparison of BSE to blood tests may not be directly relevant.
Step Two: Identifying
Context of Dialogue
An important innovation in Waltons theory is the creation of a typology
of contexts of dialogue. Walton has identified 7 distinct types of context:
persuasion, quarrel, inquiry, negotiation of interests, information seeking,
deliberation and mixed.
1. What type of dialogue is it? Is it a persuasion dialogue, a quarrel,
an inquiry, a negotiation of interests, an information-seeking dialogue,
a deliberation, or a mixed type of discussion?
This is quite clearly a species of persuasive dialogue.
2. What are the goals of the dialogue? How do they establish relevance?
The goal of this argument is to convince the reader that the conclusions
of the Canadian Task Force are not to be accepted and further that BSE
is to be encouraged.
3. Expand the analysis from point 6 of step I (above) to include the
formulation of the global conclusion. What is the issue of the dialogue?
4. Have there been any dialectical shifts during the sequence of argumentation?
Are they licit or illicit?
In this course we will not address licit and illicit dialectical shifts.
5. Is there a requirement of evidential priority? That is, do the
premises have to be established as being better known (more plausible)
than the conclusion?
In this argument, it is essential that the claims of the inadequacy of
the Russian and Chinese studies be assessed. In reading Baxters
study it is clear that the Russian and Chinese studies are not the only
studies so assessed and there is no mention of the other studies in this
argument. So, there is clearly an issue of evidential priority, because
if one can accept the premise that these studies are flawed or completely
irrelevant to the Canadian health care context, it follows that the Baxter
report should be regarded with skepticism.
6. What can be known about the positions (commitments) of each of the
arguers involved? How have these changed significantly over the course
of the dialogue?
This is more relevant for the analysis of dialogues. However, the letter
writer is the president of Breast Cancer Action, and it is a legitimate
question to ask of the organization what their official position on breast
self examination is and to what extent their organizational mandate would
be adversely affected by widespread acceptance of the Baxter report.
Step Three: Burden
1. What is the burden of proof? Give evidence from corpus.
It is not clear what the authors burden of proof is. The issue at
hand here is that the BSE report adheres to a very well developed standard
of proof. This is one of the important points of the so-called evidence
hierarchy that sets out criteria for adjudicating scientific evidence.
In advance, the process sets out what will count as evidence and what
will constitute the burden of proof.
The author quite clearly does not accept
the interpretations of the evidence. She finds the evidence is not compelling
and disagrees with the Task Forces interpretation of harms. In her account
of the evidence she leaves the impression that the two cited studies comprise
the bulk of the evidence against breast self-examination. This is misleading
and leaves the reader in a position of disadvantage.
2. Is it a deductive argument? Check for missing premises.
It is not a deductive argument.
3. Is it an inductive argument? Check for inductive errors or missing
It is not an inductive argument.
4. What types of premises are they? Are the premises statements of
fact? If yes, what is the evidentiary status of the premises? Does the
arguer state the origin of the fact? If the statement of fact is the result
of a statistical study, are appropriate ranges of uncertainty acknowledged?
Is the type of study mentioned? (See critical appraisal guidelines for
further elucidation) Are the premises statements of value or interpretation?
It is in answer to these questions that the integration of critical appraisal
and argument analysis is most evident. The argument relies on several
premises that include statements of fact.
- Allegations are made about the incompleteness of the Russian and Chinese
studies but no quantitative claims are provided to illustrate or support
- The issue about the reporting of teaching and learning can be addressed
by looking at the studies. That they were not reported could be a function
of the limitations of the journal but this can be clarified by contacting
the authors of the studies.
- How the claim of BSE causing increased anxiety is overstated is not
- The issue of increased biopsies is not analysed quantitatively, and
the 80% rate of benign biopsies is stated categorically and without
proper indication of uncertainty. (As an additional assignment, the
intrepid student may wish to track down the origin of this oft-cited
- Issues of study design are not mentioned. A common problem in health
discourse is the assumption that all studies are capable of providing
- There are distinct, but unarticulated value judgments, particularly
around social conditions in Russia and China. If Canadian women are
capable of learning BSE, then so are others.
- The author also equates the anxiety of BSE, with cancer diagnosis
and blood tests, which not all would agree with.
5. Is it a presumptive argument? Check for missing premises in appeals
It is a presumptive argument. See above for missing premises and below
for a discussion of appeal to authority.
Step Four: Evaluation
In this section, the issue of fallacious inference is addressed. We ask
that you peruse the excellent catalogue of fallacies at one of the following
web sites and then return to your argument analysis.
- Arguments against the person: There are no attacks on the person in
this argument though there is some personalization when the writer suggests
that Dr. Baxter tell women who have found lumps that BSE is useless.
- Questionable use of emotional appeals: There is an emotional tone
to the argument. Quite clearly, the writer is strongly opposed to the
conclusions of the scientific study. There are several rhetorical flourishes
indicating the authors perspective.
- Questions-asking or answering criticisms.
- Relevance. There are several aspects as alluded to above.
- Circular arguments. Some of the claims are somewhat circular. The
issue of anxiety provoked by BSE is illustrative as on average women
who performed BSE were more anxious than those who did not, so the claim
about the comforts of biopsies is partially circular.
- Appeals to authority. In the second paragraph, two Harvard doctors
are cited. Their qualifications, aside from the prestigious university,
are not mentioned. There is nothing to suggest they have expertise in
either breast cancer or cancer screening. The doctors may be well qualified,
but there is nothing offered in the argument to persuade us of this.
- Loaded definitions or terms. The analysis of harms may well be considered
- Slippery-slope problems.
- Deductive errors.
- Inductive errors.
- Questionable causal arguments.