Introduction to Philosophy
A Guide to Argument Analysis

A Guide to Argument Analysis

The analysis of arguments is key to doing good philosophy, but even more than this, it is one of the most useful tools you will take with you outside a philosophy course. Many years of teaching has shown that most students analyze an argument - which consists of both a claim (the conclusion) and reasons to accept it (the premises) -usually by just trying to establish whether some premises are false (if they want to argue against the conclusion) or by arguing that the premises are true, if they want to argue for the conclusion. Many times the arguments for or against the truth of premises are just as weak or unconvincing as the original argument, which makes for poor argument analysis. This short step by step guide to argument analysis, with tips along the way, is designed to increase your tools for analyzing arguments.

 

First things first.

Before we begin, we should define what an argument is, and what one means by argument analysis.

An argument consists of both a claim (which is called the conclusion) and reasons to accept the claim (called the premises).

To analyze an argument is to determine whether the argument actually gives good reasons to accept the conclusion, and to provide reasons why it does or does not provide good grounds to accept the conclusion.

With these definitions in mind, we will now discuss some fundamental steps in argument analysis.

 

 

Step 1. Carefully read and think about the argument

When presented with an argument in philosophy (or any other area, for that matter), you should carefully read the argument and identify the primary conclusion and all of the premises used to support that conclusion. Informally, to identify the conclusion ask yourself, What claim or point is the author of the argument trying to make? To identify the premises ask yourself, What reasons are presented to convince you of the truth of the claim? To illustrate this skill, consider this example argument:

It is a good idea to make sure you have working fire alarms in your house. Just look at the family whose house burned down on Christmas. They lost everything in their house, and almost lost their daughter who nearly died of smoke inhalation. Their house did not have working fire alarms, and for that reason, the fire itself went unnoticed while the family slept. They were saved only by the chance occurrence of a neighbor's teenage son arriving home late from a Christmas-eve date who noticed the fire and woke the family up.

 

 

 

Once we have identified the main conclusion, we can treat the remaining parts of the argument as premises used to support the conclusion.

We now pose the question, Is the above argument a good argument, in the sense previously mentioned (does it give good reasons to accept its conclusion)?

To answer this you should have determined that the conclusion is that one should have working fire alarms in their house.

What are the premises given which support the conclusion? Basically, a story is taken from the news that recounts what almost happened to a household which did not have working fire alarms. Now consider the following questions:

These questions bring out a fundamental point in argument analysis: Many times if you happen to agree with a conclusion before reading the argument, there is a greater chance that you will judge the argument for that conclusion as a good argument rather than a bad argument. This leads to the following two observations:

  1. When evaluating whether an argument is good or not, it is important to examine the actual argument itself and ignore any previous conceptions you may have about the truth or falsity of the argument's conclusion.
  2. It is possible that a conclusion is true, even when a particular argument for the conclusion does a poor job in supporting the truth of the conclusion.

 

If you are taking the online version of this course, return to your LMS (moodle, Canvas, Blackboard, etc) and complete the assignment:

writing assignment icon Creating a good argument. 

 

 

The above argument about having working fire alarms in one's house is not really "controversial" because it addresses an almost universal concern of not wanting to die of smoke inhalation, and it is argued that working fire alarms can prevent it. But other arguments we will encounter in philosophy have conclusions which run contrary to statements which you may already believe, perhaps strongly. In these cases it is very important to carefully read what the argument states and the reasons given for it. You should never argue for or against a given argument for a reason not actually given by the author! To illustrate what I mean by this statement, consider this example of an argument and a bad counter argument:

Argument: When we die, we no longer are able to think or experience anything, since experience and thinking have been shown to be the products of a working brain. Change brain chemistry or slice out parts of the brain, and we no longer think or perceive the world the same. Neuroscience has clearly demonstrated the connection between the workings of a brain and thinking and experience – just consider the many examples of antidepressant drugs or the success of anesthesia to turn off experience during a major surgery or the effect of a lobotomy, to name just a few. It is unreasonable to believe that brain processes which lead to thinking and experience continue when the brain is no longer working or functioning, which is the case when we die.

 

 

 

 

After reading the above argument. Did you agree with the conclusion before you read the argument? Did the argument give you any reason to doubt your conclusion if you did not agree, or further reasons to believe the conclusion if you did? What claims are made by the author, and what examples or evidence might the author have included? In either case, consider the following counter argument:

Counter argument. The author of the above claims that science is the only method by which things can be known, but this is not correct. Religion, meditation, and mystical experience are other ways things can be known, and for all the author or science knows some thinking might occur in the absence of any changes in the brain or brain chemistry. Since this can not be known for certain, the author's conclusion that when we die we no longer are able to think or experience anything can not be known for certain either.

 

 

 

Notice that the conclusion to the counter argument is that the occurrence of thinking and experience after death can not be known for certain. Does the author provide good reasons for this claim? Notice that the author states that the first argument claims that science is the only method by which things can be known. Is this correct? Is this really claimed in the first argument? If not, then to say such a claim was made, when it was not, does not strengthen the counter argument – if anything it weakens it, since it suggests that the person making the counter argument has not carefully considered the original. Also note that the author makes certain statements which are supposed to support the author's point of view, but gives no reason why these statements are true. For example, what evidence (or illustrative example) is given to support the claim that religion, meditation and mystical experience are indeed other ways things can come to be known? Would you be more convinced by the counter argument if evidence for this was given, or at least alluded to?

 

 

 

 

These observations lead us to the following three tips:

Tip 1. Always fairly and accurately represent the claims made in any argument you attempt to analyze. This means you actually need to address claims made in the original argument when analyzing it.

Tip 2. Don't let your conclusions stand alone! Always provide at least one reason or illustrative example to support any claim you make in your own argument. This means give a reason or reasons why any important claim made in your argument analysis is actually true. While you do not need to provide reasons for general knowledge (like Saturday follows Friday), you do need to provide reasons to accept claims which are central to any point you wish to make. For example, in the above counter argument, at least one reason (or illustrative example) should be given to support the claim that there are other ways things can be known which are not scientific.

Tip 3. When considering an argument think about additional reasons which might support the given conclusion and reasons which might not support the given conclusion. Did the author of the original argument include these? If you are arguing against the original, point out these omissions. If you are arguing for the original, point out the absence of such omissions, or offer more evidence.

In what follows we offer several other steps one can take when analyzing an argument. They can be considered in any order and used singly or all at once. Many of these steps will be new to some, and the vocabulary terms presented might have different meanings than what you might be used to, so review them carefully and take notes.

 

Step 2. Identify whether the argument is valid or invalid

Since the terms "valid" and "invalid" mean something different in the world of logic than in ordinary conversation, let us first define each and then discuss how knowing this aspect of an argument allows one to better analyze it.

An argument is said to be valid if the truth of its premises guarantee the truth of the conclusion. Another way of saying this is that an argument is valid if it is (logically) impossible to have a false conclusion and all true premises.

If an argument is valid and actually has all true premises, then the argument is called sound.

 

 

 

On the other hand, an argument is said to be invalid if it is possible to have a false conclusion and all true premises (no matter how remote that possibility may be).

Before turning to some examples, two points of common confusion should be mentioned now.

The term valid does not mean true, and invalid does not mean false. These terms describe instead possible truth conditions for an argument's conclusion, given the assumption that all of the premises are true. What this means in turn is that the terms valid and invalid describe arguments, not single statements.

And recall finally that an argument consists of a conclusion and supporting premises.

With this said, re-read the above definitions one more time.

Another common misconception is to think a valid argument has a true conclusion. This is only partially correct. A valid argument must have a true conclusion only if every single premise is actually true. On the other hand, an invalid argument need not have a true conclusion, even if every single premise is true. This information is summarized in the following table:

 

Premises

Conclusion

Argument is valid

Argument is invalid

All true

Must be true

Need not be true

Not all true (at least one is false)

Need not be true

Need not be true

 

The natural question now is how does one determine whether an argument is valid or invalid? Unfortunately a complete answer to that question goes well beyond the scope of this introductory guide. That said, one way to discover whether an argument is invalid is to just pretend for a second that all the premises are actually true – then ask yourself if the conclusion can be false without making one of the premises false at the same time. If the answer is no, then you have at least a reason to suspect the argument is valid. On the other hand, if the answer is clearly yes, then you have a sufficient reason to conclude the argument is invalid.

So how does this help with argument analysis? Well, if one knows an argument is valid, and yet suspects the conclusion to be false, the only hope one has of showing that the conclusion is false is to show at least one of the argument's premises is false. On the other hand, if an argument is valid and one wishes to argue for it, one need only provide reasons why the premises are actually true, since if they are, the conclusion must then be true.

Here is a very simple example of a valid argument:

If it is Friday, then interstate traffic will be terrible.

But interstate traffic is not terrible.

Hence it is not Friday.

Recall, we are keeping the example as simple as we can, and for the moment we are going to ignore the ability to just look at a calendar or your cell phone to check the day of the week. Now the argument is valid, but suppose you wish to argue that it is not Friday (in other words, you wish to argue against the argument's conclusion). Since the argument is valid, the only way you can do so is to somehow show either the interstate traffic really is terrible or that the connection between Friday and interstate traffic being terrible is false. In other words, you cast doubt on the second premise, the first premise, or both.

On the other hand, if you wish to agree with the argument's conclusion, since the argument is valid all you need to do is to give reasons why the premises really are true, since their truth guarantees the truth of the conclusion as that is the very definition of a valid argument.

In comparison if an argument is invalid to argue against it one can point out the conditions that would allow for all true premises and false conclusion, and show that those conditions are indeed reasonable (in the sense of "likely to be true"). On the other hand, if you wish to support the conclusion of an argument which is invalid, you need to show that the conditions that would result in all true premises and a false conclusion are quite rare. Again, we look at a simple example of an invalid argument..

If it rains, then the streets are wet.

It has not rained.

Therefore the streets are not wet.

In this case, since the argument is invalid, to argue against the conclusion all one needs to do is point out a plausible or likely scenario where the premises are still true but the conclusion is false. One such scenario might be, "It snowed last night, and warmed up today which caused the snow to melt, making the streets wet". On the other hand, if you wished to argue for the conclusion, you point out that the conditions that might normally make the premises true but the conclusion false rarely occur. For example, one might note that it is summer and hence the streets can't be wet due to snow melt. To be really successful, each reasonable possibility that would result in wet streets without rain would be addressed and rebuffed.

These observations lead to the following useful information concerning arguments.

 

Fact 1. If an argument is valid and actually has all true premises, then new information can not make the conclusion false, unless part of that new information makes at least one existing premise false.

Example. If John wins the lottery, then we will all go on a trip around the world. We did not go on a trip around the world. Therefore John did not win the lottery.

As long as new information does not change the truth of the premises, then the conclusion that John did not win the lottery stands as long as the premises are all true.

Fact 2. If an argument is invalid and actually has true premises, then new information can change the truth of the conclusion, even if the new information does not make at least one existing premise false.

Example. Thousands of airplanes take off and land without incident or crashing on a daily basis. Hence the airplane we are on will take off and land without crashing.

In this case, if one looks out the window and sees the wing on the airplane engulfed in flames, then this new information does not make the single premise (that thousands of airplanes take off and land without incident or crashing on a daily basis) false, but does lead to the very real possibility that the conclusion that this particular flight will take off and land without crashing turns out to be false.

Test you knowledge of invalid vs valid

 

Step 3. Assume all premises are true and follow the consequences

Another way one can analyze an argument is to actually consider what conclusion follows from the original premises on your own. Do you conclude the same thing as the author of the argument? Why or why not? Consider the example of our first argument given above:

 

It is a good idea to make sure you have working fire alarms in your house. Just look at the family whose house burned down on Christmas. They lost everything in their house, and almost lost their daughter who nearly died of smoke inhalation. Their house did not have working fire alarms, and for that reason, the fire itself went unnoticed while the family slept. They were saved only by the chance occurrence of a neighbor's teenage son arriving home late from a Christmas-eve date who noticed the fire and woke the family up.

Here one may argue that given the truth of these premises, one should conclude not that it is just a good idea to have working fire alarms in your house, but rather it should be illegal not to have them. Notice that concluding that "it should be illegal not to have fire alarms" is to argue another conclusion that is different and more forceful than the original which just states that it is a good idea. Since it is not immediately clear that someone who does this is arguing for the conclusion (by adding an additional condition, "not only is it a good idea, but it should be illegal not to have them), or against it, (it is not a merely a good idea, but should be illegal…), we first note the following tip and then consider two special cases:

 Tip 4. When analyzing an argument, consider other conclusions which follow from the truth of the premises which may not be included in the original argument.

 

Case 1. The reductio ad absurdum

 

The latin term, reductio ad absurdum means "to reduce (the claim) to an absurdity", where "absurdity" is meant to be taken in the logical sense of something which is impossible . This method is often used to prove a statement false. It works this way: we first assume a statement of interest is true, then by careful reasoning, where each step in the reasoning process follows validly from previous steps, if we can derive a set of inconsistent statements by this process, then the original statement must be false.

Here is an an easy example first.

cross word

Suppose you are working a crossword puzzle, and the clue for 1 across is:

Loses its cover in the fall.

You also know the letters given above for 1 across are all correct.

The clue for 1 down is:

A state of rage.

The word F R EN Z Y works.

But assuming this is true, that gives 1 across as F R E E, which is inconsistent with the clue given that 1 across is something that loses its cover in the fall.

Hence your assumption that 1 down is F R E N Z Y must be false.

 

 

Surprise! Who knew that working corssword puzzles can involve using a reductio ad absurdum!

 

Now let's turn to a more philosophical argument which uses the reductio ad absurdum.

First we will establish a miniature argument (which is nice in this case, but certainly not a normal part of a reductio ad absurdum).

Suppose you want to travel from here to some distant place in the universe. Suppose that your spaceship is really fast. But there is a catch, the distance to the place you want to travel is infinitely far away - will you ever get there? Clearly not, since no matter how long you travel or how far you go, the destination is no closer, since it is infinitely distant. Hence things infinitley distant (in space or time) are impossible to reach.

 

With this in mind we want to consider whether the following claim is true.

Any object that is not at rest was put into motion by something else.

(Think of a string of falling dominoes, if you need a picture of things put into motion by something else).

We want to use the reductio ad absurdum to examine whether the above claim is true, hence we assume it is true and follow the necessary consequences:

 

Claim: Anything that is not at rest was put into motion by something else.

Consequence 1: If the claim is true, then whatever it is which puts the object into motion must itself be in motion. We will call this object 'motion giver #1"

Consequence 2: But "motion giver #1" must also be itself in motion, and hence it was always in motion (which is impossible) or put into motion by some other object itself, which we will call "motion giver #2"

Consequence 3: But the same reasoning applies to "motion giver #2", it had to be put into motion by another motion   giver, which we will call "motion giver #3"

Consequence 4. The above sequence goes on forever or it does not. If it goes on forever then, like never being able to arrive at a destination infinitely far away, the same is true that the chain of events that stretch infinitely into the past could reach the object just placed into motion. Hence it is not true that any object that is not at rest was put into motion by something else, or there must be something that was itself not put into motion by something before it which was the object that put everything else in motion.

 

While this argument is a bit more complicated that the crossword puzzle example, it uses the reductio ad absurdum in part to establish the final conclusion stated in Consequence 4. This argument is actually similar to an argument by Thomas Aquinas in an attempt to argue for the existence of God.

But we will not go further into that argument here, and instead note that this method is also known as proof by contradiction or indirect proof. It should be noted that this method can only work when the original assumption must be false - hence it will not always work since the original assumption may not be false!

The following illustrates the reductio ad absurdum graphically:

 

A graphic showing the reductio ad absurdum method

 

Case 2. Explore the most probable consequence(s) of an assertion.

 

This method is very common in philosophy, especially so in Ethics. It is similar to the reductio ad absurdum, but instead of each step in the process being a valid inference we only demand that the new step follow with a degree of high probability. In turn, if this process leads to a conclusion that seems for independent reasons highly improbable or undesirable, then the truth of the original claim is cast into doubt. Clearly this is not as strong as the reductio ad absurdum method previously mentioned, but as that method can not always be used, this provides a useful alternative. Here is an example taken from Ethics. We will use this method on the following claim : The rightness or wrongness of an action is determined solely by the individual.

 

 

Suppose that the rightness or wrongness of an action is determined solely by the individual. This would imply that for some people, stealing might be deemed ethical, even if you personally feel it is not ethical. This in turn means that you have no justification – other than your own personal ethical code – to condemn someone who robs you and feels it is ethically permissible to do so. This consequence, namely being unable to condemn the action of someone who robbed you universally, is so undesirable in and of itself that it casts doubt on the original claim that the rightness or wrongness of an action is wholly determined by the individual.

 

Notice that the conclusion in each step is not guaranteed but seems to be likely. The end state of affairs where one is unable to condemn someone else for robbery is so undesirable that it casts doubt on the original assumption that the rightness or wrongness of an action is determined solely by the individual.

This second approach (case 2) has problems too. It may be the case that another analysis of probable inferences results in a conclusion which is probable and desirable. When this happens, all that can be done is to try to decide which chain of probable inferences is most likely. And deciding what is most likely -even with past cases as a guide - is so difficult that it even has a name and represents another problem in philosophy called the "problem of induction".

It may be the case that one attempts to argue this way, but instead of each step being likely each step is unlikely - in this case we call the reasoning process a slippery slope fallacy, which brings us to our next method of argument analysis which is an examination of logical fallacies.

The following illustrates this second method of reasoning:

 

A graphic illustrating the process of assuming the claim is true and following the probable consequences

Step 4. Identify Logical Fallacies

A logical fallacy is an argument that seems to present a good case for its conclusion, but for various reasons fails to do so. There are many logical fallacies, so many that we will only discuss one for illustrative purposes and provide a list of a few common fallacies with an online reference for further information. The point we wish to make here is that the existence of a logical fallacy in an argument does not mean the conclusion to the argument must be false or is probably false, it only means that the conclusion is not supported by the fallacious premises. This means when analyzing an argument (assuming it contains more reasons to accept the conclusion than just the fallacious ones), one should point out that part of the argument which is a fallacy, name the fallacy (or explain in your analysis why it is a fallacy), and state the conclusion of the argument is neither strengthened or weakened by the part of the argument that is fallacious.

As usual, it is helpful to illustrate this process by an example. First we need a logical fallacy.

An ad hominem fallacy occurs when some personal characteristic of the arguer is attacked rather than the conclusion of the argument itself.

Example: President Smith just signed a treaty with Bolivia which will open the doors to more trade between those two countries by removing tariffs and other economic obstacles which prevented many companies from doing business with Bolivia. However, these tariffs were put in place to prevent corporations from simply packing up and moving to whichever country is the most economically depressed and hence has the cheapest work force. The end result will be loss of jobs in the USA, and no real help to the Bolivian people, since the profits will go to those Bolivians which are already extremely rich. Beside this, President Smith's wife, who accompanied him on his diplomatic mission to Bolivia to sign the treaty, refused to eat "cuy" which is a national dish of Bolivia, offending all Bolivians present at the treaty signing ceremony. Hence not only does President Smith not care about his own fellow Americans, he has a wife who is so insensitive to the Bolivian people that she insulted all of them needlessly. Clearly President Smith's decision to sign the treaty with Bolivia is wrong.

Observe the above argument gives some reasons why the given treaty would be a bad idea, but also commits the ad hominem fallacy by attacking President Smith's wife. The correct way to analyze this argument is to point out the premises whose truth does affect the conclusion such as jobs being lost in the USA, economic benefits not going to the average Bolivian etc., and point out that the comment about President Smith's wife is an ad hominem attack and is irrelevant to the truth of the conclusion.

Since it is impossible to point out logical fallacies if you are unable to identify them, I list below just a few common logical fallacies and give a link at the end where many more are given. It is helpful to become familiar with as many as possible.

 

Some Common Logical Fallacies

Common Name

Defining Characteristics

Post hoc (ergo propter hoc)

Since y follows x, then x is the cause of y

Hasty generalization / bad induction

Takes a small set of examples and assumes all other examples are the same

Appeal to (inappropriate) authority

Appeals to the opinions of someone who is not an expert in the given field under debate

False dilemma/all or nothing/black-white

Presents two choices (many times extreme) as if they were the only ones, when there are others

Argument from ignorance

Assumes x is false since it has not been proven true, or that x is true, since x has not been shown to be false (hint: this fallacy is committed in the counter argument on page 2, can you find it?)

Slippery slope

 

A series of consequences which lead to an undesirable conclusion are said to follow from an a particular claim which casts doubt on the truth of the claim. What makes this chain of reasoning a fallacy is that each consequence is thought to be unlikely.

 

ad hominem (abusive)

 

Casting doubt or arguing against a specific claim by attacking some personal aspect of the person making the claim. Attacking the person making the argument rather than the argument.

ad hominem (circumstantial)

 

Same as above except one attacks the circumstances of the individual rather than the argument where circumstances might be; where one lives, one's nationality, where one works, one's marital status etc.

 

Begging the question/ circular reasoning

 

This occurs when the truth of the conclusion is already contained in one or more premises. In other words this fallacy occurs when a person assumes the very thing they are supposed to prove.

Strawman

This fallacy occurs when one characterizes another argument in such a way as to make it patently absurd or much easier to argue against. This fallacy is very common and also difficult to detect since its detection requires knowledge of the proper characterization of the original claim.

 

inability to be falsified

 

This occurs when a proposition (not claimed as an axiom) is such that nothing could show it to be false. In other words, normal propositions about the world have consequences if true, a proposition which has no such consequences is said to be unfalsifiable, hence to assert that it might be false is an error in reasoning.

 

Appeal to tradition

This fallacy occurs when the truth of a claim is said to follow from traditional views or claims like "that is the way it has always been . . .".

ad populum / appeal to the masses

This fallacy occurs when the truth of a claim is stated to follow from the mere fact that everyone believes it to be true or, equivalently, no one doubts it.

 

The above list is quite short. There are dozens of such fallacies. Thankfully there are several great online sites which list many more together with examples.

See the sidebar Learn more for links.

 

 

 

 

 

Conclusion

Notice that in all of the above steps, not a single one starts first by examining the actual truth or falsity of the premises – which is normally the first (and perhaps only) method that beginning learners use when starting to analyze an argument. Instead many of our methods focus on the nature of the inference and what follows if we assume the premises are actually true.

Finally we state last which should be stated perhaps first, as it is as important as any of the tips we offer above:

 

Final tip: One should decide the reasonableness of a given conclusion in light of all available arguments. It is bad reasoning to decide beforehand whether you accept an argument's conclusion before careful consideration of arguments for and against and the relevant evidence.

 

In closing we remind the reader that the above steps are not exhaustive. There are many more ways one can analyze an argument (for example, the very powerful method of argument by analogy or the role of definitions in argument analysis), and much more could be said about the actual steps that we have outlined. But when learning something new, it is best to start out with small steps and then move onward, and hopefully the steps presented here accomplish that task.

 

 

Quiz

 

Please return to your LMS (e.g. Canvas, Moodle, Blackboard. . . ) and complete the quiz Guide to Argument Analysis.