Argument strategies are various ways we present our arguments and justifications. Some arguments are simple deductions and generalizations based on our experiences. However, there are a variety of other argument strategies, and a better understanding of them can help us learn to argue more effectively. Argument strategies are usually compatible, and we can often present our justifications using a variety of argument strategies. I will discuss four argument strategies and give examples of these strategies used within the philosophical literature :
- Argument from analogy
- Thought experiment
- Argument from absurdity
- Inference to the best explanation
1. Argument from analogy
An analogy is when we compare two different things to emphasize a relevant similarity between those two things. For example, both kicking and punching are often morally wrong because they are intended to hurt people and often succeed in doing so. We could say that kicking and punching are “analogous” insofar as they are both similar in a certain way and are often morally wrong due to that similarity.
Not all analogies are good. Some of them are “false analogies.” Many people even argue that “all analogies fail.” Is the analogy drawn between kicking and punching a false analogy? Someone could argue that kicking and punching is a false analogy because it’s wrong to kick people in a boxing match, but it’s not wrong to punch people in a boxing match. However, this objection is unconvincing. Many people seem to assume that analogies are meant to prove that two things are “equivalent” but that isn’t the purpose. It is true that kicking and punching aren’t equivalent. They are two different things. However, I think the analogy succeeds in revealing that they are similar in at least one important respect—they are both often wrong when they are used to hurt people. Whenever it’s wrong to hurt someone in ordinary contexts, it’s wrong to punch or kick that person.
When are analogies false? When they fail to have the relevant similarity. For example, some people have suggested that legalizing same-sex marriage is analogous to legalizing marriage between a man and dog, and they are both wrong for the same reason. However, marriage to a dog is morally wrong because dogs can’t consent to the marriage. Two men can both consent to marry each other unlike a man and a dog.
An example of a philosophical analogy was given by Peter Singer in “The Drowning Child and the Expanding Circle.” He argues that a professor going to class who spots a child drowning in a shallow pool of water and can save the life of the child at minimum cost to herself has an obligation to save that child—and we have a duty to give to charities that can save lives at a minimum cost to ourselves for the same reason—because we have obligations to do a great deal of good when doing so is at very low cost to ourselves. The professor who can save the child and everyone else who can give to an effective charity are in analogous situations insofar as they can both save lives at a minimal cost. Some people have objected to Singer’s analogy by arguing that the professor doesn’t have an obligation to help the child or by citing that there’s some important difference between the two situations. It’s not obvious that Singer’s argument succeeds, but it is philosophical insofar as it’s thought provoking and has been a rich source of debate.
2. Thought experiment
Thought experiments are imagined situations that are meant to illustrate a point. Thought experiments can be used to illustrate analogies, help us realize when a belief is intuitive (or counterintuitive), or prove a theory to be inconsistent.
Not all thought experiments prove what we would like them to. An example of a misused thought experiment is Aristotle’s assumption that heavier objects fall faster than light ones, perhaps just by imagining it in his mind, and then concluding that it must be true. Such a thought experiment might prove that it’s intuitive to expect heavier objects to fall faster and many people share this intuition, but it’s false that heavier objects fall faster than less heavy ones.
Singer’s analogy used to prove that we have an obligation to give to charity was illustrated by a thought experiment. Other examples of philosophical thought experiments include the following:
- Hilary Putnam’s “Twin Earth” – Imagine that another world doesn’t have H2O and instead has XYZ (another chemical) that functions exactly like water, and all their experiences of XYZ are exactly like our experiences of water. It quenches thirst, it looks and feels wet, it boils at 100 degrees Celsius, people call it “water,” and so on. Putnam argues that XYZ is not water, and therefore “meanings” are not merely in our heads. Instead, the meaning of our words can have a component based on the nature of the world itself (and perhaps a causal connection or history of our words). When I say “water” I’m referring to the stuff in our world (H2O) because of the history involving the stuff I drink and call “water” throughout my life. “Twin Earth” illuminates how intuitive it is to think that water is H2O and no other chemical combination (no matter how such a chemical functions in our lives).
- John Searle’s “Chinese Room Argument” – Imagine that you perform the same tasks as a computer that seems to “speak Chinese.” A woman gives you a piece of paper with Chinese letters on it, you follow a set of instructions to write something in Chinese and give it back to her. What you write back makes sense. Do you know Chinese? Searle answers, No. Computers follow instructions just like you do to “speak Chinese” without actually understanding the language. Searle argues that this reveals to us that no amount of instructions is sufficient to understand a language—the meaning of the words and sentences (the semantics of a language). Searle’s “Chinese Room Argument” emphasizes that we have little reason to think that we can understand semantics from following instructions alone because it’s intuitive to think we wouldn’t in the thought experiment he presents us with.
- Judith Jarvis Thomson’s “Defense of Abortion” – Imagine that people kidnap you and hook you up to another person to keep that person alive, and you alone can keep that person alive by staying connected. You could keep yourself connected to the person for nine months to keep her alive, or you can disconnect yourself and let her die. Are you obligated to stay connected to her? If people’s “right to life” overrides our rights to our own bodies, then yes. However, Thomson thinks not, and she argues that women (who become pregnant from rape) aren’t obligated to stay pregnant for the same reason—the right to life doesn’t override a person’s right to their body (when we never chose to be responsible for another person’s life). Thomson’s thought experiment illustrates an analogy—between a far-fetched scenario and a women who wants an abortion (caused by rape).
- Frank Jackson’s “Epiphenomenal Qualia” – Imagine that Mary is a scientist who studies color but never experiences color. She lives in a black and white room with a black and white television, black and white books, and so on. She knows everything involved with color involving wavelengths of light and neurobiology. However, she will still learn something if she gets a color television and sees what color looks like. Jackson argues that this thought experiment is evidence that qualia—the “what it’s like” element of our experiences—can’t be known entirely from a description of causal processes. Jackson’s thought experiment illuminates our experience of qualia and emphasizes that we have little reason to think a description of causal processes could tell us everything there is to know about experiencing qualia (because it’s intuitive to think it couldn’t within the thought experiment).
3. Argument from absurdity
The argument from absurdity or “reductio ad absurdum” is a strategy used to (a) provide evidence against a belief or argument or (b) prove that something is true when denying it leads to absurdity. This is done by assuming an argument is sound (or belief is true) and showing the absurd consequences of doing so. These absurd consequences are often “counterexamples”—states of affairs that would be impossible if the argument was sound (or belief was true).
How can we use the argument from absurdity to object to a belief? We could assume the belief is true and show that it requires us to reject another belief we can’t rationally deny. For example, someone could claim to know that nothing is morally wrong, but we might argue that “if that’s true, then there’s nothing morally wrong with torturing a small child, but we know there is something morally wrong with that.” The fact that we know that it’s wrong to torture children is a counterexample to the belief that nothing is morally wrong.
How can we use the argument from absurdity to prove an argument is invalid? We can assume that the reasoning used by an argument is valid (it can’t have true premises and a false conclusion at the same time) or effective and show that true premises can lead us to a false conclusion. If a form of reasoning can use true premises to prove something false, then we know the form of reasoning to be invalid and unreliable. Consider the following two examples:
Someone could argue:
- All humans are mammals.
- All humans are animals.
- Therefore, all mammals are animals.
This could sound like a good argument because the premises and conclusion are all true, but it’s actually an invalid form of reasoning that could be used to give us false conclusions (even when the premises are true). The argument form is: “All A are B, all A are C, therefore all B are C.” A counterexample will have to use this form of reasoning and have true premises, but A, B, and/or C will stand for something else. For example:
- All humans are mammals.
- All humans are primates.
- Therefore, all mammals are primates.
We know that all mammals aren’t primates because dogs and cats are mammals, but thy aren’t primates. Merely changing the word “animals” to “primates” was enough to create this counterexample. The fact that this argument form can have true premises, but a false conclusion proves that it’s invalid and unreliable.
- If something is legalized, then people will probably become more tolerant and want to make more things legal.
- Therefore, if we legalize same-sex marriage, then the next thing you know we might legalize marriage to nonhuman animals.
- It’s wrong to legalize marriage to nonhuman animals.
- Therefore, we shouldn’t legalize same-sex marriage.
In this case the first premise is questionable and how exactly the second premise can be derived is unclear. A good counterexample to this argument can show that it makes use of the “slippery slope” fallacy by showing an analogous argument that makes use of the slippery slope fallacy in a very similar way. For example:
- If something is legalized, then people will probably become more tolerant and want to make more things legal.
- Therefore, if we legalize owning guns, the next thing you know we might legalize owning tanks.
- It’s wrong to legalize owning tanks.
- Therefore, it’s wrong to legalize owning guns.
This counterexample reveals how absurd it is to expect certain causal effects based on legalizing various behavior. The fact that we legalize same-sex marriage or ownership of guns is unlikely to have the terrible consequences that are discussed. This analogy rests on the assumption that there’s no significant difference between the two scenarios. Some countries and civilizations have legalized same-sex marriage without legalizing marriage to nonhuman animals, and countries that legalize the ownership of guns hasn’t lead to the legalization of owning tanks. We have no reason to expect these things to happen.
The reductio ad absurdum is traditionally used in logic and mathematics as an “indirect proof” to show that a form of reasoning is valid by assuming it’s invalid (the premises can be true and the conclusion can be false at the same time), then showing why it’s impossible for it to be invalid because assuming it is leads to a contradiction. For example, the following argument uses valid reasoning:
- If Lassie is a dog, then she’s a mammal.
- Lassie is a dog.
- Therefore, Lassie is a mammal.
We can assume the conclusion is false in an attempt to derive a contradiction and prove that the argument is valid. For example:
- If Lassie is a dog, then she’s a mammal.
- Lassie is a dog.
- Lassie is not a mammal. (Assumed premise)
- Therefore, Lassie is not a dog. (If Lassie was a dog, then she would be a mammal, but we are assuming she’s not a mammal.)
- Lassie is a dog and she is not a dog. (This is a contradiction.)
- Therefore, it’s impossible for our original argument to be invalid. (The assumed premise lead to a contradiction.)
In logic the indirect proof tends to be written with placeholders (letters) to abstract away from the content and make it clear that we only want to prove that the form of reasoning is valid rather than that a certain premise is true. The argument form is “If A, then B; A; therefore B.” In order to prove this is valid, we can assume it’s invalid. Let’s assume the premises are true and the conclusion is false at the same time. We can assume the conclusion (B) is false by assuming “not B.” If that’s impossible because it leads to a contradiction, then we know it’s valid:
- If A, then B. (Premise)
- A (Premise)
- Not B (Assumed Premise)
- Not A (“If A, then B” and “not B” proves “not A” because the truth of A would prove the truth of B.)
- A and not A (This is a contradiction.)
- Therefore, the original argument was valid. (If the premises are true, then the conclusion must be true, and the assumed premise must be false.)
The indirect proof is not so different from the other forms of the argument from absurdity that I discussed. For example, if we wanted to prove that some things are right or wrong, we could assume that nothing morally wrong, and then we can explain how we know it’s wrong to torture children, so it’s impossible for our assumption to be true. This is the same argument as given above without technically being an objection to a belief anyone accepts.
Failed arguments from absurdity
Not all arguments from absurdity succeed. For one thing we sometimes assume two beliefs to be incompatible that aren’t actually incompatible. Someone might try to prove that evolution is false because of a “counterexample”—such as the argument that evolution must be false because evolution says we evolved from fish, but fish still exist. (Or that we evolved from monkeys, but monkeys still exist.) However, there’s nothing about evolution that guarantees that all animals of one type will simultaneously evolve into another type. It’s possible for one group of fish to evolve into something else without all fish evolving into something else. Imagine that human beings colonize Mars and end up stranded there for a million years. By the time the Earthlings meet up with the former-Earthlings again, they could both have evolved into differing creatures by then. The former-Earthlings might evolve into green people, like how we might imagine Martians to look; and the Earthlings could have evolved into furry blue people with cat-like ears.
4. Inference to the best explanation
An inference to the best explanation (also known as “abduction”) requires us to consider all the viable explanations (and contrast them) to find out which is most likely true. We can consider various strengths and weaknesses of each explanation to find out which is best. This is an important part of the scientific method—scientific hypotheses are explanations for a phenomena and scientists use the best hypothesis (or theory) that’s available. Such a hypothesis is often very plausible and many scientific theories seem to be proven to be very accurate, such as Einstein’s theory of relativity.
An inference to the best explanation isn’t enough to prove that everyone should hold a certain belief because it might not be entirely clear how likely it is that an explanation is best, it might be that rival explanations are also plausible, and it might be that the best explanation available is itself implausible. Nonetheless, an inference to the best explanation could be a helpful argument even in these situations.
Two common forms of inference to the best explanation involve (a) identifying the possible existence of various things and (b) identifying possible causes. Scientists inferred that disease was often caused by small invisible organisms (germs) before we could ever see them using microscopes. This inference included both the possible existence of an entity and the possible causal effects of that entity. The fact that doctors who sterilized their instruments and washed their hands helped prevent infection provided evidence for the hypothesis that germs exist before we could see them.
The conclusion that all men are mortal seems like a very good example of an inference to the best explanation, which is based on our expectation that if any men are immortal, we would have found out about it by now, and people would spread such a discovery to make sure it’s well known. There are alternatives—perhaps immortal men keep to themselves and don’t stay in one place for too long in order to make sure they are never discovered. We simply don’t think this is a likely possibility.
Some people argue that an inference to the best explanation is “inductive reasoning” rather than “deductive” because the conclusion isn’t guaranteed to be true by the evidence, and the conclusion is only (relatively) probable. However, this is true of almost all philosophical arguments and inductive reasoning could be taken to be one type of deductive reasoning with a premise that states that we can generalize using a sufficiently large sample. For example, the inductive argument that all men are mortal based on our reasonable expectations, can be written as the following deductive argument:
- If we don’t know of any immortal men, then the best explanation is that there aren’t any (because we have reasonable expectations that if any men are immortal, they would probably be discovered and we would know about them, and we find alternative explanations far-fetched).
- We don’t know about any immortal men.
- Therefore, the best explanation of our observations is that there aren’t any immortal men.
This is a deductive argument that uses the argument form known as “modus ponens” (If A, then B; A; therefore B.)
We must be careful when we think we make an inference to the best explanation because we could misjudge the plausibility of the alternative explanations. For example, the inference that a ghost must have moved my keys because they aren’t where I remember leaving them is not an inference to the best explanation because there are alternatives that are more plausible. For example, I probably just misremembered where I left them.
An example of a philosophical argument for the best explanation is that we are justified to believe that other people have minds because it’s the best explanation for their behavior. It’s possible that (a) we are deceived by mindless bodies that behave as though they have minds, (b) that there’s an elaborate hoax involving machines that look like people, or (c) that my entire life is a dream and no one else exists. However, all of these alternative explanations for the behavior of other people seems far-fetched. Other people seem to have a very similar biology to myself and it seems reasonable to think that similar biology leads both to similar minds and behavior guided by mental activity.
Almost every philosophical argument can gain strength through an inference to the best explanation because it requires us to consider “both sides” (or all viable sides) of an issue. For example, Peter Singer’s argument that we have an obligation to assist is based on the assumption that people will share his intuition that we have an obligation to help save lives when doing so comes at a low cost to ourselves, but we might wonder if this intuition is a reliable form of evidence. It’s possible that some people can know that we have an obligation of this sort—perhaps from personal experience of some sort—but is this the best explanation of our intuition? Some people might worry that it’s more likely that we are merely prejudiced due to indoctrination to think that we are obligated to help others when doing so costs us little. More will need to be said to convince a skeptic that such an intuition is reliable.
It is important to have sufficient evidence to reach our conclusions, but it’s also important to present our evidence in a reasonable way that makes it clear how we reach our conclusion. Argument strategies are ways of presenting our evidence in a reasonable way. Every argument strategy can be misused. There are false analogies, misused thought experiments, failed arguments from absurdity, and failed inferences to the best explanation. That doesn’t mean that these strategies are always misused. They are all common forms of reasoning and seem to be effective a great deal of the time.