Category: Research

Getting there from here: Critical thinking and debiasing

Cate Hundleby

A primary lesson from the first day was that biases pose different problems in different contexts and need specific techniques to be negotiated. Ways to engage the social and emotional variables occupied a central role on the second day. In that afternoon, and today, the third and final day, we get more specific, considering how reasoning operates in particular contexts and the difficulties faced by critical thinking pedagogy in addressing those contexts and improving people’s practice.

Scientific reasoning concerns Ulrike Hahn, who directs us toward the advantages of Bayesian approaches to fallacies and argumentation. Hahn’s point of departure comes from Douglas Walton‘s analysis of argumentation schemes, including taking the appeal to ignorance as a central case. Accounting for how we gain and incorporate information into belief is the key benefit of Bayesian probablistic analysis, and Hahn makes quite a convincing case, at least if we take arguments to be monological. She takes the dialogical approaches of Walton and pragma-dialectics to become unnecessary in the wake of the Bayesian appraoch. Burden of proof and its dialogical shifts are a messy business, to be sure, but I don’t think we can dispense with analyzing them so easily. We need tools to reason with others, and judging by the social approaches to debiasing that prevailed on the second day, the social dimensions may be key to debiasing argumentation.

Mariusz Urbanski presented research completed with Katarzyna Paluszkiewicz and Joanna Urbanska showing that while untutored skill has a lasting effect on performance in deductive reasoning, training can improve performance on difficult deductive problems.

Debiasing was studied in legal contexts by Frank Zenker, Christian Dahlmann and  Farhan Sarwar, because it’s an environment that demands judgements be made; judges can’t suspend their evaluation of a case. People, and so perhaps judges, who score high on agreeableness seem especially influenced by anchoring.

This day gave us little room for optimism about debiasing, or the future of critical thinking as an ideal and an educational practice. An account of frustration with facilitating critical thinking at an institutional level provided the final talk, from Chip Sheffield, recounting his experience as the inaugural Eugene H. Fram Chair in Applied Critical Thinking at Rochester Institute of Technology.IMG_1510 Sheffield described how institutional pressure to develop or adopt a new assessment instrument flew in the face of a lack of other institutional and faculty support, student resistance to the integration of critical thinking in courses, and only superficial resources in the larger critical thinking movement. Yet he found some progress in student panels and discussions, and community outreach.

The result of Sheffield’s experience is his recommendation of more critical thinking courses, and more at advanced levels. So it might seem that the problem is not how to teach critical thinking but how much to teach it. That could seem to be the only room for improvement given the disjunct between the experimental research on cognition and biases and the philosophical pedagogy of critical thinking. I was struck to hear the experimental psychologists express concern over the philosophers use of “bias” losely, just as the philosophers were discouraged by the inattention to the pressures to find practical strategie for students.

However, as much as this conference demonstrated an absence of the practical evidence we want to guide critical thinking education, it also paves the way for more engagements between the psychological evidence and the pedagogical and philosophical ideals. I take particular inspiration from the work of Kenyon & Beaulac, and Howes, as philosophers attempting to explore the implications of the psychological research, but also Weinstock, Schendel, Cooke, plus Urbanski, Paluszkiewicz & Urbanska, and Zenker, Dahlmann & Sarwar in assessing the impact of specific techniques on the development of critical thinking. We have barely begun to chart the possibilities and potentials for critical thinking education, and we need more research of the sort gathered at this conference to develop and explore options. Ideally, this should be like the work from Paglieri, Mercier & Boudry, directly engaging both the philosophical and the psychological. To start with, we should explore whether they or Goodwin are correct about the effects of classroom debate.

RACT watch this space

Cate Hundleby

IMG_1348-0 Tomorrow begins the conference on Reasoning, Argumentation and Critical Thinking Instruction at Lund University in Sweden. Dr. Frank Zenker, a researcher at Lund, has hosted several conferences in recent years on social epistemology that I would have loved to attend, but I’m especially happy to attend this one as it deals with an intersection of academic topics of special importance. It’s an open question whether argumentation fosters critical thinking and how it affects reasoning, and the recent decades of psychological research on reasoning suggest that people’s thought is much less critical than we might tend to otherwise recognize. All sorts of bias — from perceptual to social — colour our interaction with and reasoning about the world.

The picture at the top is of a lovely brass band playing in the pub this evening, and here are a few speakers who’ve convened so far.

IMG_1355 I will post summaries and commentary on the talks as we proceed over the next few days.

What Is Argument Repair?

Cate Hundleby

Somewhere along the way, I picked up the language of “argument repair.” I don’t think this is my term, but borrowed from others. Many people have found it the most interesting concept in my analysis of textbook treatments of fallacies.  Some textbook authors have asked for sources, and some instructors have developed their own exercises.

I have only found one scholarly discussion of it, and no articles aside from mine, but in critical thinking and argumentation studies it’s quite common for  innovative scholarship to be presented in textbook form.  (A great recent example of this is Maureen Linker’s Intellectual Empathy: Critical Thinking for Social Justice — I plan to post on that soon.) Richard Epstein’s Critical Thinking (Thomson Wadsworth) devotes a chapter to the argument repair, and perhaps that’s where I picked up the term.  It’s no longer in print, unfortunately, but you can find a discussion of it here.

What attracts me to the notion of argument repair is that it suggests a dialectical context. We may repair arguments in response to criticism, though we cannot in change our central reasoning (how to identify that central reasoning is the big question). That dialectical engagement distinguishes argument repair as a practice from simply charitable interpretation, which does not involve the arguer revising what is said but only an audience interpretation. (It may also include repairing one’s own argument, in an internal dialectic.) That dialectical element is present when “argument repair” appears in the one other place I’ve found the phrase, a poem “On the Immortality of the Soul” by Charles William Jones (of Islington):

In matters of such import, pith, and weight,
As the soul’s welfare and a future state,
Each should endeavour for himself to judge,
Invoke the contest, and refuse to budge:
Then to the field of argument repair,
Wisely engage in manly conflict there;
But should conviction’s force your breast assail,
Retire at once, nor your defeat bewail

This rich depiction of argument as war is far from what I want, and yet conflict is certainly part of the dialectical picture.  What interests me is that the audience allows the speaker the opportunity for argument repair.  The speaker can take “to the field” and “retire” in turns.

Although dialectical context distinguishes argument repair from simply charitable interpretation, each may follow the same guidelines.  We don’t want to shoehorn all arguments into deductive form, and yet we want to allow for implied premises, and also “blunders” in Douglas Walton’s terms (see Tindale’s article here).  Epstein in a later edition along with Carolyn Kernberger provide the following Guide to Repairing Arguments (206, 68):

Given an (implicit) argument that is apparently defective, we are justified in adding a premise or conclusion if it satisfies all three of the following:

  1. The argument becomes stronger or valid
  2. The premise is plausible and would seem plausible to the other person
  3. The premise is more plausible than the conclusion.

If the argument is valid or strong, we may delete a premise if doing so doesn’t make the argument worse.

The three criteria involve additions to the argument: the audience “putting words in the speaker’s mouth” in a non-pejorative sense, or the speaker making additions; but it’s also noted that an argument might be improved but trimming off extraneous information. Epstein and Kernberger also suggest the following limits to argument repair (206, 68):

  • There’s no argument here

  • The argument is so lacking in coherence that there’s nothing obvious to add

  • A premise it uses is false or dubious and cannot be deleted

  • Two of its premises are contradictory, and neither can be deleted

  • The obvious premise to add would make the argument weak

  • The obvious premise to add to make the argument strong or valid is is false

  • The conclusion is clearly false.

None of this indicates a dialectical context, so it’s interesting that they choose terminology different from the traditional “charity.”  So, it seems that I need to look closer at my own past practice to see if I’m doing anything different, and have made explicit my interest in the dialectical context. I’ve also sent out some inquiries to people who might know better than I, and I’ll report back.

Critical Thinking and “Academically Adrift”

Cate Hundleby

The central purpose of my CT2 guide is to advise philosophy instructors about their choices in developing critical thinking courses.  The responsibility for the development of critical thinking skills does not rest solely on instructors however, as one can see in Richard Arum and Josipa Roksa’s 2011 book, Academically Adrift: Limited Learning on College Campuses.

Other important factors include student preparation, work habits, and social resources — both work and funding. For instance, educated parents, studying alone rather than in groups, and working on campus improve performance on the Collegiate Learning Assessment  (chapter 2 and chapter 3).  All this, they suggest, indicates trouble for the desired goal of increasing “critical thinking,” complex reasoning, and clear written communication, though these terms are never defined. Part of the problem according to Arum and Roksa is the current culture of college education that presents studying as part-time, and school as a last fling before the sombre (and sober?) world of work (chapter 4).

Nevertheless, Arum and Roksa conclude with recommendations for instructors. They advise more reading and writing, with the caveat that it may not go over well with students and may look bad on student evaluations and other currently accepted measures of teaching.  How much?  40 pages of reading per week and 20 pages of writing per term. This seems a lot to me.  I assign about 30 pages of reading per week (10 per class hour), based on small books, and so often as little as 15 pages per week.  I maintain, and explain to students that philosophy must be read three times to be understood well, and that our reading demands thus will be less than in history and English.  The same goes for writing, I’d argue.  Few sources are needed, so the analysis is more spare and focussed and I could not expect it to be as long as in other disciplines.  But now, I’m just getting defensive … I imagine Arum and Roksa might accept this ammendment.

What they really want is for us to engage in what they consider a more direct ethic of education, to produce reasoners rather than credentials.  They turn explicitly to the tradition of John Dewey in stressing rigour and frequent assessment.  This would entail a turn away from the collaborative learning models that have become popular. Again, this might hurt teaching evaluations, and would demand deep structural change to support instructor-level practices.

I have a number of misgivings about the book, and many criticisms can be found on-line.  My biggest concern is that the results can seem to show simply that individual reasoning does not benefit from collaborative instruction.  That seems unsurprising.  The need for assessment to reflect instructional methods and techniques is known as “alignment” and — as I understand it — the canonical work in this regard is Teaching for Quality Learning at University by John Biggs.  Ignoring this element deeply undermines Arum and Roksa’s approach as the technique is central to the outcomes based model of education they criticize.  It’s akin to giving a US-based IQ test to an African student and remarking at how poor the results are.

That said, if we agree that the skills assessed by the CLA are important, we need to consider how those outcomes have been side-lined in favour of more concrete and specific, and — yes — social skills.  Do we want individual or collaborative thinkers? Collaboration is more popular with students but that is that what employers want?  I haven’t yet seen this general question posed to Arum and Roksa, though some concerns have been similar.  They admit that they do not account for the subject-specific strengths students may gain.  But I’m pointing to something a little different, an understanding of general reasoning as collaborative or cooperative, which conflicts with the enlightenment notion of reasoning implicit in the CLA.  That type of reasoning is important, but there are alternatives.

What’s wrong?

Cate Hundleby

The main motivation for the Critical Thinking Squared website and this blog is to raise consciousness about critical thinking pedagogy in the discipline of philosophy.  There are a few ways in which scholarship has advanced in recent decades that textbooks continue to ignore.  Many of these advances come from argumentation theory, and here are the basic advancements in that field that I find textbooks tend to neglect:

  1. Arguments take place in dialogues.  Some theorists view the dialogue as a secondary aspect; but the consensus is that among the range of forms that argumentation can take, arguments take place between people.  They are not free-standing but audience-relative, an insight going back to Chaim Perelman and Stephen Toulmin.
  2. Fallacies are forms of argumentation. Without this assumption there is little hope of providing any coherent account of fallacies. While fallacy labels provide a longstanding method for teaching reasoning, the errors we count as fallacies take many different forms and until the development of the informal logic movement there was little method behind the lists of error names.  The depth of this problem has been set forth by Charles Hamblin in his book Fallacies (1970).  Since then, fallacies have been systematized in three different ways:
    1. The three aspects of an argument that may go wrong (acceptable premises, relevance of premises to conclusion, and sufficiency of premises in supporting the conclusion).  This analysis structures  the classic books by Johnson & Blair and Govier.
    2. Presumptive argumentation schemes that may work well or go awry.  Identification of the proper working of a presumptive scheme of argumentation, such as an appeal to authority or a generalization, one considers critical questions characteristic of the particular scheme.
    3. The pragma-dialectic approach treats fallacies as rule violations but that provides no systematic account of how they arise.

Many textbooks employ the first analysis, but many still lean on ad hoc lists of names and view most types as categorical problems despite the flourishing of epistemologies showing the relevance to reasoning of emotion, testimony, generalization, etc., all of which challenge such categorical dismissals.