Thinking Critically

“There are two ways to slide easily through life:

to believe everything or to doubt everything.

Both ways save us from thinking.”

- Alfred Korzybski

The Watson-Glaser Critical Thinking Appraisal is a famous critical thinking test.

Watson and Glaser define critical thinking as a combination of:

1. An attitude to inquiry that involves the ability to recognise problems and an acceptance of the need for evidence in support of claims.
2. Knowledge of the rules and restrictions of logical reasoning.
3. The skill to apply this attitude and knowledge.

So how do we do this then? First, we take the attitude that nothing is going to get past us. We are the bouncers on the door of debAte, the members’ club where all the coolest arguments go. If they’re not on the list, they’re not coming in. How do arguments get on the list? They have to satisfy these criteria:

1. All claims are backed up with reasoning or evidence or both.
2. The conclusions follow logically from the premises (i.e. the argument is valid).
3. The premises are true (provably, or at least there is evidence for them being true).
4. There are no redundant or stuck-on conclusions.

Let’s go through those one-by-one:


1. All claims are backed up with reasoning or evidence or both.

If you make a claim, whether factual or emotional, you need to back it up. The most common mistake on Debatewise is making unjustified claims.

You can justify your claim with reasoning or with facts, but using both is better.

Example:

“The sky is blue.”
- unjustified claim

“The sky is blue. This is clear from the fact that when we look up at the sky, we see blue.”
- reasoned claim (the reasoning is flawed, however!)

“The sky is blue because of ‘Rayleigh scattering’. As light moves through the atmosphere, most of the longer wavelengths (red end) pass through. However, much of the shorter wavelength light (blue end) is absorbed by gas molecules. This absorbed blue light is then radiated in every direction, getting scattered all across the sky. Whichever direction you look, some of this scattered blue light reaches you. So you see a blue sky. [ref1][ref2][ref3]”
- reasoning and three separate references (yes, you are allowed to use Wikipedia for one of them!)

This is also something to look for in other people’s arguments. Seeing whether someone has backed up their claim will help you to evaluate how strong their argument is.


2. The conclusions follow logically from the premises (i.e. the argument is valid).

There is a detailed tutorial on the logical mistakes that can occur in arguments on this blog (How to Win an Argument, parts 1-4).

However, if you don’t want to read 7,000 words on logical fallacies, here is a short introduction to how logic can help you think critically:

Non sequitur is Latin for ‘doesn’t follow’. This describes an argument where what someone claims doesn’t follow from what they started with. This might be because the argument is invalid or unsound (we’ll look at ‘soundness’ in the next section).

Most arguments say something like “because of A and B, therefore C”. A and B are called premises and C is called the conclusion. The ‘validity’ of an argument is just based on its structure, the ‘soundness’ is based on whether the statements are really true.

Valid means:
If the premises are true, then the conclusion is true. Always.

Invalid means:
If the premises are true, that doesn’t necessarily mean the conclusion is true (it could be true, but that would just be a coincidence).

A valid argument:
All men are mortal.
Ben is a man.
Therefore, Ben is mortal.

A valid argument:
All animals with wings can fly.
Penguins have wings.
Therefore, penguins can fly.

The second argument is definitely valid, but its conclusion is false - penguins can’t fly. How has this happened? The first premise “All animals with wings can fly” is not true. Notice that ‘validity’ is only about the logical structure of the argument, not about whether any of the statements are really true. The first argument above is sound, the second one - because its first premise is false - is unsound.


3. The premises are true (provably, or at least there is evidence for them being true).

So, we’ve seen how even valid arguments can produce false conclusions - because they are unsound:

Sound means:
A valid argument with true premises (and so a true conclusion).

Unsound means:
A valid argument with at least one false premise OR an invalid argument.

(If an argument is invalid, it is automatically unsound.)

We could argue that “All animals with wings can fly” is actually true. This would stop the argument being unsound and so mean that is was a good argument. To do this, we would need to provide evidence or reasoning (see point (1)) to show that it was true.

If we were against this argument, we would only need to show ONE animal with wings that could not fly to show that the whole argument is unsound. A penguin, for example!

As well as using things that are wrong to show that an argument is unsound, you can use unsound arguments to prove that things are wrong. This is called reductio ad absurdum, which means ‘reducing to an absurdity’.

For example, if your opponent claims that “All animals with wings can fly”, then you can use the argument above to conclude that penguins can fly (which is absurd) and so show that your opponent’s claim is false.


4. There are no redundant or stuck-on conclusions.

All of these four criteria for good arguments can be difficult to spot or difficult to use. Redundant or stuck-on conclusions are probably the most difficult.

Imagine an argument is trying to get into the members’ club debAte. He’s got good credentials, this argument, but he’s got a few friends with him. If we let him in, should we let his friends in without checking their credentials? No. If an argument appears to conclude two different things (or more), then we need to check each conclusion separately.

Here is an extreme example to illustrate:

“1+1=2” and “2+2=4” are true statements.

We could say something like:

If “1+1=2”, then “2+2=4”
1+1=2
Therefore, 2+2=4
---- This argument is valid and sound.

But because “2+2=4” is ALWAYS true, that means that this next argument is also valid and sound:

If “1+1=2”, then “2+2=4 or the moon is made of cheese”
1+1=2
Therefore, 2+2=4 or the moon is made of cheese

So we have managed to conclude that the moon being made of cheese is somehow related to 1+1=2 and might actually be true!

Keep a watchful eye out for these stuck-on conclusions; most of the time they are not even known to the person giving the argument.

---

So, a good argument needs to satisfy the entry conditions:

1. All claims are backed up with reasoning or evidence or both.
2. The conclusions follow logically from the premises (i.e. the argument is valid).
3. The premises are true (provably, or at least there is evidence for them being true).
4. There are no redundant or stuck-on conclusions.

How to read an argument

1. Get the feeling.

Read the argument from an emotional point of view. Try to feel what the author wants you to believe/understand/agree with. Does the argument feel persuasive? Do you feel like they might have a good argument?

Consider how reliable the author’s words feel as you re-read the argument. Do they sound like they know what they are talking about, or are they just launching into opinionated claims? Do they sound like they have vested interest in the subject, or are they arguing from an objective standpoint? The author’s reasons and psychology are important when considering their argument.

If there are any references, it is worth reading them to see how reliable the sources are. Wikipedia or newspapers can be reasonably reliable, but it’s usually better if they’re supported by other, more reliable sources.


2. Get the main points.

The next step is to draw out the main points of the argument. This involves going through it a few times to brush away the fluff and filler and find what is actually being said. Is the argument coherent (does it make sense to go from one step to the next)? Is the argument consistent (does it ever disagree with itself)?

Has the author written hundreds of words when one or two sentences will have done? Are they waffling and distracting from the weakness of their argument with a volume of words? Or do they actually have a lot of very important stuff to say? What about the other way round? What if they make a good point, but have only said it in one sentence and then never explained it?

A very useful thing to do here is write down the main points and see how strongly each one is justified. A lack of justification is a common mistake on Debatewise.


3. Go Spock.

When we get very logical about the whole thing, I like to call it Going Spock. This can be very useful when you’re up against an argument that seems unassailable. By picking logical holes in an argument, you can find its weaknesses and exploit them.

Being really logical requires a decent insight in to various logical fallacies and a basic understanding of the rules of logic (some of which can be counter-intuitive). These things are not needed, however, to just be very rigorous with your analysis. Don’t be afraid to be pedantic!

Go Spock!

---

Let’s look at some examples from Debatewise:

Patents on life-saving drugs should be bypassed (15 September 2010)
- From the Irrational to the Ingenius[sic]:The Status Quo and Our Proposal

No, because...
The current patent system is flawed to its very core. Under the status quo pharmaceutical firms make use of research from state-funded university labs, that has been released into the public domain, to base their own research programs. They usually enter late in the process when there already is a possible product in sight. Once a product is completed it is patented for the sole use of the company for a period of 20 years. This 20 year monopoly is rigidly enforced by both domestic & international law across the world.

Devastating problems arise with this because the high prices charged by these firms to maximize their profit are too expensive for those in developing countries, and even in some developed countries that lack proper social security institutions, resulting in millions of unnecessary deaths. The current system suppresses innovation by placing legal hurdles to further researching and improving existing drugs. It also incentivizes spending on non-life threatening diseases common in the West (such as hair loss) and therefore bring in more revenue than those focused on poorer countries that can’t afford to spend as much on medication.

Therefore, we propose a plan set out by Economics Nobel Laureate Joseph Stiglitz where an international multi-billion dollar fund would be set up to award scientists and firms once they discover or develop cures to life-saving diseases. This would aim to cover research costs but also incentivize research into life-threatening diseases over other ailments. Once payment is made the product is pooled into the public domain, available for use by any company anywhere in the world. While this would be expensive (0.6% GDP) it would pay-off in the medium term as health care systems would no longer have to pay huge premiums to drug companies. More importantly, it would allow for millions to afford medicines [1].

[1] http://www.independent.co.uk/opinion/commentators/johann-hari/johann-hari-the-hidden-truth-behind-drug-company-profits.html


1. Get the feeling - corporate = bad, big pharma = bad, solution = utopian system; author is reasonably credible, but makes a lot of unjustified claims. The only reference is from a newspaper and it is a broken link; it would have only been one journalist’s opinion of a scientific report - it is better to have read the report yourself and use that as a reference.

2. Main points - current system is flawed: too expensive and prioritises non-life-threatening diseases; they propose a solution based on a suggestion by an academic.

3. Going Spock - where is the justification for the all the claims made about the current system? A reference would help, but evidence would be better. Where are the facts that say hair loss is prioritised? The solution, while delightfully labelled with some academic’s name, isn’t analysed to show its effectiveness at solving the problems listed. So it is presented as a solution, but it isn’t explained how it solves. Finally, and this is getting quite pedantic, the current system being flawed is based on whose ideal? What ideal system are they comparing it to, to say that it is “flawed to its very core”?

---

Is the UK education system getting better? (15 September 2010)
- School is to educate !

Yes, because...
Seeing as the point of school is to educate students and a better result would suggest they have been educated better which would mean the education system is getting increasingly better.
[link to graph of results going up 1988-2010]

Let’s just jump straight in and go Spock on this one:
What are they actually saying?
A. The point of school is to educate.
B. Better results suggest that students have been better educated.
C. Results are getting better.
D. Therefore, the UK education system is getting better.

(A) is irrelevant to the argument about the education system being better or worse. (B) and (C) are premises and (D) is a conclusion. Does the conclusion follow from the premises? No. The conclusion that would be valid is (E) “Therefore, students have been better educated.” This is similar, but not the same as (D). Also, (B) may be true, but is not necessarily the whole truth. There may be other things that improve results.

This is noticed by the opposition:

No, because...
A better result could be caused by something other than better teaching: easier exams or the way examinations are marked not working properly, children being naturally more intelligent, or being motivated by the recession to work harder in order to gain qualifications needed for jobs, parents paying more attention to their children's educations.

We could add that students being better educated does not necessarily mean that the UK education system is getting better (i.e. (E) does not imply (D)).They could have got better another way.

---

Remember - Don't be afraid to be pedantic. Make sure you are rigorous in your analysis. 'Leave no stone unturned.'