CRITICAL THINKING: WHAT IS IT?

Robert H. Ennis
University of Illinois, UC

In their interesting and stimulating papers Professors McCarthy and Norris analyze critical thinking from different standpoints. McCarthy attempts to justify the teaching, non-teaching, testing and non-testing of various aspects of critical thinking at various times. Norris urges the empirical study of critical thinking topics. Each paper has strengths and weaknesses. I shall elaborate.

McCarthy, using for the most part an ordinary language approach, distinguishes between episodic and dispositional terms, suggesting that ‘critical thinking’ is only an episodic term, while ‘critical thinker’ and ‘rationality’ incorporate dispositions (which are not episodes; they are tendencies to behave in certain ways in episodes). She attaches pedagogic significance to this distinction, holding that we can teach and test episodic critical thinking, but that we neither can nor should attempt to teach or test for dispositional rationality.

Although McCarthy is correct in identifying this episodic sense of ‘critical thinking,’ the word is also used as a label for an area of curricular concern — at least an ability or body of abilities that we attempt to teach. Abilities are not episodes. It is similar with the word ‘reading.’ The word ‘reading’ can be applied to a particular episode, as in “I am reading this paper,” but we do also try to teach reading in school, just as we try to teach critical thinking.

In fact her limitation interferes with one of her points. She wants to teach “the episodic ‘critical thinking,’” but that does not make sense. You do not teach episodes. You teach people how to do well, and to do well, in episodes. That is, you teach skills, abilities, dispositions, and the like.

McCarthy also deals with Harvey Siegel’s attempts to justify critical thinking. If I understand her correctly, I agree with her points, and made similar ones in an exchange with Siegel several years ago.¹

Furthermore, McCarthy argues against the direct teaching of critical thinking dispositions. She holds that one approach, charismatic modeling, is indoctrination, and thus is inappropriate. The question, then, is that if this is indoctrination, what is the matter with indoctrination (of this sort)? The topic, indoctrination, has a long history in the annals of philosophy of education. McCarthy makes no use of this extended discussion of the topic. The topic needs elaboration here.

Another possible difficulty with her attack on the direct teaching of critical thinking is that one of her recommended methods seems to be direct teaching. This method is to point out to students the advantages of critical thinking — and the real life consequences of failing to think critically. This seems to me like direct teaching. Possibly McCarthy’s idea of direct teaching needs further examination.

McCarthy’s favored method of promoting critical thinking dispositions is to teach the subject matter well and deeply, pointing out consequences of failing to think critically. Then one waits, perhaps for years, trusting the student to realize the value of a disposition and to adopt it. Because of this developmental process, testing for dispositions would, she thinks, be useless; the students at the end of a course might not yet have developed their dispositions. For the efficacy of this extreme laissez-faire approach to teaching she offers no evidence. She just says, “It would seem that” to support her claim.

Even so, there are several problems. First of all one has to wonder which subject matter she wants taught. Many of my decisions that have real life consequences (and that called for the exercise of critical thinking dispositions) are not in any particular subject matter that I have ever studied. One example is my decision as a juror about the guilt of a defendant in a murder trial. This topic has been discussed elsewhere at some length by John McPeck and me.² McCarthy only touches the surface.

A second problem resides in the obvious cases of people who are experts in their fields — who know their subject matter very well — but who are lacking some crucial critical thinking dispositions. Cyril Burt, the renowned psychologist who falsified data, is an example. He is only one of many people who are well informed but lack certain critical thinking dispositions. Thorough knowledge of the subject matter is not enough.

In her discussion of rationality McCarthy — without argument — claims that morality implies rationality. That is, if a person is moral, the person is rational. It would be interesting to know why she thinks this. I know someone quite well who is often not rational, but seems quite moral to me. Discussion of a case like this and an argument in support of the position would help to illuminate the topic.

McCarthy does offer an argument against the converse proposition she mistakenly attributes to me, the proposition that if an action is rational it is moral. Her case of the wise stock market investment seems to be an effective counterexample. However I did not state the proposition she challenged. What I said was that if an action is rational, it is not immoral. There is a significant difference between “moral” and “not immoral.” An action that is not immoral could be morally neutral, which incidentally is how she characterized the stock market investment. Her example is consistent with what I said.

In sum McCarthy discusses a number of topics, too many to do justice to any. Interestingly her failure to supply any evidence to support her empirical claim about teaching subject matter is in conflict with the spirit of the other presentation in this symposium, Stephen Norris’ paper, to which I now turn.

I am in strong general agreement with Norris’ emphasis on the need for empirical research in this field. But I am not happy with his treatment of the topic, the definition of critical thinking.

Norris provides us with three types of nouns, nominal-kind terms, strict natural kind terms, and non-strict natural kind terms. His examples in turn are “bachelor,” “buckyball,” and “gander.” He suggests that Harvey Siegel’s “critical thinking” is a strict natural kind term (like “buckyball”); that John Mcpeck’s “critical thinking” is a nominal-kind term (like “bachelor”); and that my “critical thinking” is a non-strict natural kind term (like “gander”).

He urges that empirical research is needed to find the denotation of ‘critical thinker,’ or the underlying traits of the critical thinker. He sees this research as the activity of a philosopher-empirical researcher partnership committed to identifying referents for the term, ‘critical thinker.’ A term that does not refer he believes to be useless in studies that attempt to improve education.

Since Siegel and McPeck are here to explain themselves, I shall concentrate on Norris’ classification system and its application to my definition of ‘critical thinking.’ However I suspect that my concerns apply to Norris’ handling of Siegel and McPeck as well.

My basic concern is that Norris focuses on descriptive terms (apparently assuming that all statements are either analytic or synthetic). However, ‘critical thinking’ is a value-laden theoretical term, an “essentially contested concept.” His examples are all low-level descriptive terms. Even though some folks would say that even low-level descriptive terms are also value-laden terms, they are not value-laden to the extent that empirical research cannot determine their denotation or the underlying traits — by Norris’ own account. But consider terms like ‘good’ and ‘justice.’ Can empirical research determine their denotations, or the associated underlying traits?

We need another category, which I shall name “concept-conception terms”³ to fit terms like `good,’ `justice,’ and `critical thinking.’ The meaning of the term `good’ is captured by the OED definition, “a term of general or indefinite commendation.”⁴ But empirical research will not tell us what is then denoted by that term. Rawls defines the concept justice roughly as fairness.⁵ But empirical research can not then tell us to what the word `justice’ refers — simply on the basis of its definition as fairness, because it cannot tell us what is fair. That is, empirical research cannot give us a conception of justice, a set of rules for identifying cases of justice. That is what Rawls tried to do in his book, Theory of Justice. Similarly I define the concept critical thinking as “reasonable reflective thinking focused on deciding what to believe or do.”<sup>6</sup> I attempt in this definition to reflect the central tendency of usage of this term in English-speaking North America, a project similar to those of the OED and John Rawls. But empirical research cannot then tell us who would be denoted by `critical thinking’ in accord with this definition. How can empirical research identify what is reasonable?

What I have tried to do with my extended list of abilities and dispositions⁷ (to which Norris referred) is to offer — with much guidance and help from others — a set of guidelines (that is, a conception) that I feel will help us reach reasonable decisions. The guidelines cannot be determined by empirical research. Furthermore they cannot be applied by empirical research. Experienced well-informed human beings must use their judgment in applying them.

Consider the guidelines for judging the credibility of a source: A source is credible roughly to the extent that is has expertise, has no conflict of interest, agrees with others equally qualified, has a good reputation, uses established procedures, knows that its reputation is at risk if the statement is refuted, is able to give reasons, and is careful. These guidelines are not necessary conditions and must be weighted in accord with the person’s appraisal of the situation. How can empirical research make these decisions in order to decide whether the person is to that extent a critical thinker? Not possible. So not even this conception, much less the concept, of critical thinking can be applied by empirical research.

Empirical research can do important things. It can count how many people do and do not satisfy the guidelines in various topic areas, given a judgment about each individual’s satisfaction of the guidelines (as Deanna Kuhn⁸ did, and as a number of social psychologists have done and are doing, described in part by Nisbett & Ross⁹) It can run correlations with other variables. It can get statistical summaries of test results (assuming that the test score is ultimately determined by human judgment applying a conception). But it cannot by itself identify the referents of either the concept or conception of critical thinking that I offered. It can only help to organize the judgments required for such identifications.

In sum, even though I have much sympathy for Norris’ general demand for more empirical research about critical thinking issues, I agree with John McPeck on this issue. Identifying the referent of the term, ‘critical thinker,’ is not the task of the empirical researcher in the way in which Norris wants.

McCarthy and Norris have each offered us stimulating provocative commentaries on the current critical thinking scene. But McCarthy’s work needs more elaboration and empirical backing, and Norris wants empirical research where it does not fit.

---

¹ Robert H. Ennis, “The Rationality of Rationality: Why Think Critically?” in Philosophy of Education 1989, ed. Ralph Page (Bloomington, Illinois: Philosophy of Education Society, 1990), 402-405.

² Robert H. Ennis, “Critical Thinking and Subject Specificity: Clarification and Needed Research,” Educational Researcher 18 no. 3 (1989): 4-10; John McPeck, “Critical Thinking and Subject Specificity: A Reply to Ennis,” Educational Researcher 19 no. 4 (1990): 10-12; Robert H. Ennis, “The Extent to Which Critical Thinking is Subject Specific: Further Clarification,” Educational Researcher 19 no. 4 (1990): 13-16.

³ I owe this distinction to John Rawls in his Theory of Justice (Cambridge, Massachusetts: Harvard University Press, 1971), 5.

⁴ Oxford Universal English Dictionary (Oxford: Oxford University Press, 1937), 811.

⁵ More precisely he says, “Institutions are just when no arbitrary distinctions are made between persons in the assigning of basic rights and duties and when the rules determine a proper balance between competing claims to the advantages of social life.” Rawls, 5. How can empirical research decide whether rules are “arbitrary” and whether we have achieved a “proper balance?”

⁶ “A Taxonomy of Critical Thinking Abilities and Dispositions,” in Teaching for Thinking: Theory and Practice, ed. Joan B. Baron and Robert J. Sternberg (New York: W. H. Freeman, 1987), 10. Also “A Streamlined Conception of Critical Thinking,” Teaching Philosophy, 14 no. 1 (1991), 6.

⁷ In the items mentioned in previous footnote.

⁸ Deanna Kuhn, The Skills of Argument (Cambridge: Cambridge University Press, 1991).

⁹ Richard Nisbett and Lee Ross, Human Inference: Strategies and Shortcomings of Social Judgment (Englewood Cliffs, New Jersey: Prentice-Hall, 1980).