February 27, 2009

Some Critical Thinking Skills Are Critical

This is a long one kids, so bear with me. There is a payoff in the end.

Wikipedia provides a reasonable definition of critical thinking:

Critical thinking is purposeful and reflective judgment about what to believe or do in response to observations, experience, verbal or written expressions, or arguments. Critical thinking might involve determining the meaning and significance of what is observed or expressed, or, concerning a given inference or argument, determining whether there is adequate justification to accept the conclusion as true.

Critically evaluating assertions, arguments, and proposals, whether presented orally or in print, is an important comprehension skill. Many personal, professional, and social decisions are based on what we are told by other people. Because faulty arguments and propaganda are so common, critical thinking has a role in almost every important decision we make.

The consensus is that the typical K-12 education does not result in students with good critical thinking skills. I agree with this consensus.

What I do not agree with; however, is the notion that critical thinking is some generalized skill that is independent of domain knowledge.

Let's see why this is so by looking at skills needed to critically read a passage of text. I am going to use the procedure outlined in Direct Instruction Reading, 4th edition, chapter 22, which is a simplified version of the skill suitable for elementary school students, but, which sadly most students never acquire.

The four steps in the critical-reading process can be treated as the major component skills:

(1) identify the author's conclusion; that is, what does the author want the reader to believe?

(2) Determine what evidence is presented; that is, what does the author present to convince the reader? Evidence or opinion?

(3) Determine the trustworthiness of the author, that is, can the reader trust what the author says?
  • Does the evidence come from a qualified person?
  • Does the person have biases?

(4) Determine if the conclusion derives from the evidence. Identify any faulty arguments.
  • Tradition, either old or new (sometimes called a bandwagon effect)
  • Improper generalization
  • Confusing correlation with causation (or coincidence)

Those who think critical thinking/reading is a generalized skill are confusing the general procedure with the act of performing the procedure itself. A student might have learned the procedure, but be unable to perform the procedure adequately for a particular text. That's where domain knowledge comes into play.

The best way to see this is by way of example. So, let's use the procedure to critically read a typical passage that students might encounter.

Thomas Edison, the inventor of the lightbulb, was seriously concerned about the increasing use of alternating current as a form of electricity. Edison believed that because alternating current involved so much more current than direct current, alternating current was a threat to the nation. Many fires were caused by alternating currents. In fact, alternating current was used in Sing Sing to electrocute criminals. Direct current was used with light bulbs for many years. Edison felt direct current was still the best form of electricity.

Another example of the dangers of alternating current has just occurred. A house wired with alternating current caught fire and burned to the ground. The fire started when an electrical wire became so hot that a wall caught fire. Alternating current will eventually cause a fire whenever it is used. Direct current rather than alternating current should be used for lighting.

The Daily Post used direct current to light its press room for over a year. Reporters are much happier now. They write more interesting stories. Sales of the newspaper have increased dramatically. The Daily Post is now the most popular newspaper.

Identify the Author's Conclusion

First the students must use details from the passage (seriously concerned, a threat to the nation, direct current is still the best) to form a main idea (or author's conclusion). Identifying an author's conclusion is a continuation of summarization skills.

Right off the bat you can see that critical reading/thinking is dependent on domain knowledge and the student's ability to extract the author's conclusion from the text and justify that conclusion with supporting details from the text. A student might be able to extract the main idea from Dr/ Seuss's Green Eggs and Ham, but be unable to extract the main idea from one of Plato's Dialogues. If critical thinking/reading were a generalized skill, then this wouldn't be the case. The student's inability to identify the author's conclusion will impact his ability to perform the remainder of the procedure.

Discriminating Evidence from Opinion

The second step is for the student to decide whether the author's conclusion is based on opinion or evidence. If it is based on opinion, students must understand that the conclusion is nothing more than a suggestion by the author about what people should think. A conclusion based on opinion does not imply that the student should believe or act on it.

In the example passage, both opinion and evidence are used to support the author's conclusion. The statement that alternating current is a threat to the nation is opinion. The other details are evidence used to justify the author's conclusion (occurrence of fires, electrocution of criminals, initial use of direct current as an energy source.)

This step is also intertwined with domain knowledge. Being able to discriminate between fact and opinion requires that you understand the underlying statement in the first place.

Fact or opinion?

In my judgment, the total enthalpy of any non-isolated thermodynamic system tends to decrease over time, approaching a minimum value.

It's not an opinion, even though it uses a phrase ("In my judgment") that typically indicates opinion. It is evidence. The statement is the second law of thermodynamics. But, you wouldn't know this unless you knew thermodynamics, and, thus, you wouldn't be able to discriminate between opinion and evidence.

In actuality the statement is inaccurate evidence. The statement is not the second law of thermodynamics; I changed some of the terms to make it inaccurate. Again, you wouldn't know this unless you knew thermodynamics.

Determining the Trustworthiness of an Author

The third step consists of several questions, all relating to the reliability or trustworthiness of the person presenting the argument.

Question (a) is whether the evidence comes from a qualified person. Since Edison was definitely an expert on electricity in the late 1800s, he was qualified.

Again, this determination requires that you know something about Thomas Edison. That's domain knowledge.

Question (b) concerns biases the expert might have. In Edison's case, two major biases existed. One was his deep personal and involvement in a company that provided direct current. He stood to lose money if alternating current replaced direct current. Also, his reputation was at stake. He became famous, in part, because of his discovery of the lightbulb and a distribution system for electricity based on the use of direct current. If alternating current replaced direct current, his reputation might be diminished.

Again, we need domain knowledge knowledge to make the determination.

Since Edison's biases contribute to the passage's conclusion, the evidence he cites may not be trustworthy.

Since there is doubt about the trustworthiness of the author, students must seek information from different experts. The statement that direct current is the best form of electricity is disputed by many experts. Alternating current can be transmitted great distances, but direct current cannot (at least during the relevant period). If remote areas are to receive electricity at a reasonable rate, alternating current is a necessity.

So, since the expert is biased and alternate interpretations of the evidence are compelling, the evidence is probably not trustworthy.

Again, the student must know quite a bit about electricity and its ability to be transmitted to make seek out the appropriate sources and understand the information presented. take a look at this article on electric power transmission and see if you can make sense of the differences between direct current transmission and alternating current transmission. And then you have to be aware of the history of electric power transmission to understand that the information this article presents was unknown during the time that our passage was written. But, how would you know this without quite a bit of domain knowledge in both electricity and the history of electrical power transmission? Google may be your friend, but he's not this good of a friend for you.

Identifying Faulty Arguments

The final step in the critical-reading process is deciding whether a conclusion legitimately derives from the evidence. In many arguments, valid evidence will be presented, but then a conclusion will be drawn that does not derive from the evidence. In the alternating-current example, one possible interpretation is that since direct current has been used with lightbulbs for many years, it should continue to be the best form of electricity. This faulty argument illustrates the use of tradition: what has been the best must continue to be the best. Conclusions based on tradition are not necessarily true. What has worked well may continue to be the best procedure, or a better procedure may be developed. Students can disregard conclusions based on tradition. (Note that the same attitude can be taken toward newly developing traditions; i.e., "Everybody is starting to use alternating current; therefore, you should, too." A conclusion that a product or procedure is better because it is popular is faulty.)

In the second paragraph, there is an example of improper generalization. One valid example is presented, but then a conclusion is drawn that applies to all examples. One fire caused by alternating current does not mean that alternating current will cause a fire every place it is used. Improper generalization occurs often: "I saw a rich person who was rude. What makes rich people rude?" "We sat next to a long-haired man in the movies. He smelled. I'll bet he hadn't bathed in weeks. Long-hairs should take better care of their bodies."

The third paragraph involves a confusion of causation and correlation. An event that is associated with success or some other positive outcome through coincidence is erroneously concluded to be the cause of the positive outcome.

Direct-current lighting is associated with happier reporters, more interesting stories, and greater sales; however, direct current did not necessarily cause reporters to be happier. The electric lighting that produced the positive outcomes could have been achieved with direct or alternating current. Conclusions suggesting causation that are, in fact, based on correlation can be disregarded. Confusion of correlation and causation is often made: "Joe Blow uses Squirt-Squirt deodorant, and girls always chase him." "Sally took You-Bet-Your-Life vitamins every day. She lived to be 106."

Of course, there are many other logical fallacies and faulty arguments that the student should know, but these three are a good start for primary grade students since they are among the most common.

Sadly,this last step is often completely ignored in most schools. Most students are simply not taught how to identify faulty arguments at all. And let me be crystal clear here: it should be taught. And, I'm guessing that the reason why it isn't taught more often, or at least learned by most students, is because it is the culmination of basic reading instruction which is rarely reached by many students.

Many students never learn how to decode accurately and proficiently in grade-level texts. This is criminal, quite frankly, because this is something we have known how to effectively teach for some time now. Then students need to comprehend what they've decoded well enough to extract the main idea and the justifications therefor. Most students can't do this competently when the main idea is presented to them as an alternate in a multiple choice format, much less being able to generate one on their own. And, if a student can't do this initial step in critical reading, how is she going to accomplish the accomplish the remaining steps which require an understanding of the author's conclusion in the first place?

So, what we have is two separate problems. One problem is that students are often not taught how to spot faulty arguments and logical fallacies which are important skills needed to think and read critically. No one seriously disputes that this is a problem. These skills should be taught. Period.

But, even if students are taught these skills, they will not necessarily be able to think and read critically because knowing how to identify and distinguish faulty arguments and logical fallacies is not the same thing as critically thinking/reading as many proponents of 21st Century skills seem to think. Critical thinking/reading involves much more than identifying faulty arguments, as I described above. And, the ability to identify faulty arguments is not the generalized form of critical thinking as some seem to think.

There is another problem that must also be dealt with. That problem is that students can't comprehend well enough to extract the author's conclusion, to discriminate evidence from opinion, and to determine if the author is trustworthy. All of these skills require and are a function of the student's domain knowledge. You have to know a lot of stuff, to think about a lot of stuff.

This is just a roundabout of saying that there are prerequisites to being able to think/read critically that also need to be addressed (and have never been adequately addressed) before students can make use of the "21st century skills" many think are important, like being able to identify faulty arguments and logical fallacies.

What's the sense of teaching a student astrophysics if the student doesn't know how to do basic arithmetic or understand basic science? Perhaps this is why schools haven't traditional taught these things in the first place.


Stephen Downes said...

> Right off the bat you can see that critical reading/thinking is dependent on domain knowledge and the student's ability to extract the author's conclusion from the text and justify that conclusion with supporting details from the text.

You have not established this, and the (cherry-picked? created?) example does not prove your point.

Critical thinking texts (including, presumably, the one you cite) are clear that there are some *general* techniques that are commonly used (a) by authors to indicate premises and conclusions, and (b) by readers to identify them.

For example, a widely used method is the use of the indicator word. An indicator word identifies either a premise. If i use the word 'thus' or 'therefore' or even 'so', for example, this indicates that a conclusion will follow.

Similarly, my use of the words 'because' or 'since' indicates that a premise will follow.

These indicator words allow a reader to learn the nature and form of the author's reasoning, even if they have no domain-specific knowledge.

Indeed, (and crucially), it is impossible for a person to learn domain-specific knowledge (beyond the mere recitation of data) without comprehending and employing indicator words.

In addition to indicator words, a number of other techniques are used to identify premise, conclusion, and form of reasoning.

AGAIN: cherry-picked examples do NOT establish an argument. That you continue to argue this way, as though you are actually making the point, mystifies me.

KDeRosa said...

Critical thinking texts (including, presumably, the one you cite) are clear that there are some *general* techniques that are commonly used (a) by authors to indicate premises and conclusions, and (b) by readers to identify them.


These indicator words allow a reader to learn the nature and form of the author's reasoning, even if they have no domain-specific knowledge.

Your counterexample is only applies in the limited situation where the indicator words are present and dispositive. So at best, your argument only applies to a subset of argument.

Yes, indicator words can often help the reader extract the conclusion.

So, there is the simple situation where the use of indictor words is dispositive and can be relied upon by the reader to extract the conclusion.

Then there's the situation where indicator words are not used at all and can't be relied upon to extract the conclusion.

The use of explicit indicator words is hardly universal. In many situations, the reader, on his own, must extract wholly or partially the conclusion from the text.

And thenn we have the situation where domain knowledge is still required for basic comprehension of the text in the first place.

But, of course, you know this.

Indeed, (and crucially), it is impossible for a person to learn domain-specific knowledge (beyond the mere recitation of data) without comprehending and employing indicator words.

You haven't established this now have you?

It's also patent nonsense.

Indictor words are often omitted in texts. often the main idea is the first sentence of a paragraph, in which case, it often is presented as a declarative sentence without an indicator. In fact, texts are often written as a series of declarative sentenecs (except perhaps in critical reasoning books designed so that students can easily identify premises and conclusions).

Let me adopt your own conclusion: cherry-picked examples do NOT establish an argument. That you continue to argue this way, as though you are actually making the point, mystifies me.

Anonymous said...

Occam's razor can beneficially be applied to the term "critical thinking."

There are substantive terms/concepts that define domain knowledge. And there are reliable protocols for examining arguments, which your post explains. There are also sometimes subtle matters of syntax and semantics involved, such as Downes' example of
indicator words."

Each of these matters can and should be taught, which I take to be largely your point. But tnen concluding that some reified abstractions termed "critical thinking skills" are "critical" is a nice play on words, but is counterproductive to your argument. Or so it seems to me Ken.

I don't see how anyone could disagree with you that the distinction between valid and faulty reasoning is very different from from domain to domain--say baseball, law, physics--or pick any example you like.

There are other matters worth teaching. Identifying reified abstractions is important locically and scientifically. Identifying metaphorical frames of reference per George Lakeoff. Propaganda and advertising techniques. The protocol for identifying Ponzi schemes could have saved prominent investors billions. And it would have been nice for everyone had we known how "complex derivatives" work. The rules for taxonomic frameworks. Rules for validating scientific theory. System and organization matters. This is the stuff that subject matter, disciplines and domains are made of. I certain agree that they deserve much more instructional attention than they have received.

But none of these matters are specific to "reading" and terming them "thinking" doesn't add anything either.

Anonymous said...


I think there are five types of skills conflated under the label "critical reasoning."

1) Basic cognitive skills such as recitation (Accurately recalling a sentence contained in working memory)

2) More advanced cognitive skills such as completing syllogisms or detecting faulty logic

3) High frequency vocabulary (such as the word "judgment")

4) Domain specific knowledge generally assumed by most well written textbooks (usually called Core Knowledge). This is knowledge that is (sometimes) needed to learn from textbooks or read mass media critically.

5) Low frequency domain specific knowledge needed to critically analyze poorly written texts or intentionally misleading texts (Propaganda.)

These skills need to be clearly broken out, as one reason they are not often taught is that educators conflate them.

Another reason they are not learned is that teaching ABOUT these skills is different from teaching MASTERY of these skills. For example, college students are no more "logical" in their writing after completing a college level course in logic. Teaching elementary students ABOUT the concept of Author's Bias is different from teaching to mastery the skill of Detecting Bias.

KD, you seem to be making a case that what we generally refer to as "critical reasoning" is impossible without low frequency domain specific knowledge (category five.)

The first four categories are general skills within reach of any school system and should be promoted, as they are rarely taught and almost never taught well.

Students who are faced with unfamiliar content can use skills in the first four categories to obtain the relevant domain knowledge. For example, they could do some extensive reading about Thomas Edison. Google is also a better friend than you give it credit for. Just google "history of electric power transmission" or read the wikipedia link that pops up.

It will always be less work to read critically in a field in which you are not an expert, but it is possible. And the developing the general critical reading skill required to do so is an important educational goal.

Anonymous said...

Yikes, Anon.

That's a creative construction of a "hierarchy of critical thinking." But you have no base.

KDeRosa said...

Dick, I am also not a fan of the oft-abused term "critical thinking," but the wikipedia definition seems accurate enough to me. Though I certainly do not think that this is the layman's understanding of the term. Anyway, that's why I used the term.

I used the term "critical reading" because that's what the procedure was called in the DI Reading book because the skills were being taught in the context of reading.

As you can see, I waffled between the terms since I really do not like either. I'd prefer something something along the lines of threshold information acquisition skills but which is both cumbersome and inaccurate.

Anon, your definition of category 5 is way too narrow. Every piece of information is potentially biased, untrustworthy, and rife with faulty arguments.

You also seem to be arguing that students can learn domain knowledge on the fly as they need it. I at least partially disagree, the successful googlers also seem to possess quite a bit of general knowledge already which makes googling productive for them. Smart people make the mistake in assuming that everyone is as smart as they are. take a look at Donn Hirsch article on why you just can't look things up.

Anonymous said...

He became famous, in part, because of his discovery of the lightbulb...

I hate to nitpick, but Edison did not discover light bulbs (unless, perhaps, there were a box of them in hidden away in the lab that he'd forgotten about and then found). Edison invented light bulbs.

Feel free to correct the error and delete this comment.

Anonymous said...

Here's another thing to throw in the mix...

I'm an engineer, electronics by degree, but close enough to have a working knowledge of electricity on a grand scale. When I read the piece there was no need to extract the intricacies of logic or syntax. From the start, my domain knowledge was screaming out for mercy. I had a hard time believing this was a seriously written passage that was being quoted. Was it?

My point is that the amount of critical thinking skill required seems to be inversely proportional to the domain knowledge of the reader. Isn't that the point, after all?

With only moderate domain knowledge one would need more critical thinking skill than the domain expert. If you have a tin ear you won't spot a song that's off key.

Here's a rhetorical question to ponder. If your life depended upon a correct analytical decision, would you submit the problem to a highly skilled syntactic logician with 20th century skills honed to perfection, or would you put it in the hands of a hard chargin', rough around the edges, whiskey drinkin' domain expert?

Tracy W said...

Paul B - I trained as an electrical engineer (though I never actually worked as one) and also as an economist. Part of my work involves translating between the two groups. My domain knowledge has screamed out for mercy quite regularly, albeit mostly in the economics field, for example, engineers and physcists have a massive tendency to ignore that money must go somewhere, so if the government spends money supporting some industry to maintain jobs that means that it must either raise taxes or borrow money or spend less on something else, which likely will reduce jobs. (There are more sophisticated arguments for assistance to at least some industries.)

However if you want to read stupid things written about energy (which I think any electrical engineer can laugh at) go look up energy healing sometime.

My point is that the amount of critical thinking skill required seems to be inversely proportional to the domain knowledge of the reader. Isn't that the point, after all?

I think that Ken's point was that as your domain knowledge grows critical thinking in tht domain becomes far easier. Like most of us can now walk without thinking about it, after as small children spending ages developing this skill quite consciously. Or in other words, we are using a lot of skill, but we get so good at it we're not aware of how good we are.

NMahoney said...

This is addressed to Downes and others of his ilk. Would you please use your critical thinking skill set to judge the validity of this opinion:

"That's one way, but you'd have to somehow instruct the compiler to use that for every pointer dereference. The easier method is to go in and change the values during compaction. Compaction is also known as stop-and-copy; it starts with a live set, everything you can reference from the stack, then copies over only the live objects while modifiying every pointer that uses it. It's messy but it works. Allocation is dead simple and fast. There's no fragmentation. And the runtime is limited by the live set rather than the heap size. There is a huge downside, however.

I wouldn't recommend mixing anything resembling C pointer maths with compaction, since its incredibly difficult to tell what's a pointer and what isn't (in fact, without modifying the compiler, it can't be done in C or C++). For this reason, the Boehm collector (a collector that replaces new and delete) goes for the Mark-and-Sweep method instead of compaction. Because you dont move objects, you don't have to worry about figuring pointers. Boehm's collector is also called conservative, not only because it doesn't modify live objects, but also in that it treats any data on the stack or in the heap as a potential pointer. If the data points inside the heap, the object containing that address is marked. This can lead to false positives on occasion, but there's no helping that without any support from the compiler (again contradicting the grandparent). The good news is that a false positive isn't going to cause direct harm in mark and sweep. All that happens is that space that could be used isn't; Boehm claims this is irrelevant in today's operating systems with virtual memory, although I doubt you'd see an entire page's worth of false positives. Certainly, I can't do any better than him."


Without suitable expertise in the subject matter, being armed to the hilt with critical thinking skills will do virtually nothing for most lay readers of this passage.

People turn away and turn off when they have no idea what the text is even talking about. It's insurmountable. You seriously think otherwise?

KDeRosa said...

Excellent example.