Toggle menu
122
332
11
3.4K
Information Rating System Wiki
Toggle personal menu
Not logged in
Your IP address will be publicly visible if you make any edits.

Informal logic and fallacies

From Information Rating System Wiki
Revision as of 16:28, 8 September 2024 by Pete (talk | contribs) (Created page with "{{Main|Logic}} <span id="informal-logic"></span> == Informal Logic == Most informal logic is based on analyzing arguments, written in plain English, for known fallacies. Fallacies can occur in all types of arguments (deductive, inductive, abductive) but in deductive arguments a fallacy is regarded as an invalid argument, equivalent to a math mistake which can be established using the rules of inference and equivalence discussed before Internal:FromGitlab/Deductive_&_...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Main article: Logic

Informal Logic

Most informal logic is based on analyzing arguments, written in plain English, for known fallacies. Fallacies can occur in all types of arguments (deductive, inductive, abductive) but in deductive arguments a fallacy is regarded as an invalid argument, equivalent to a math mistake which can be established using the rules of inference and equivalence discussed before Deductive & Inductive Arguments and Use Cases:

  1. All dictators are corrupt.
  2. All politicians are corrupt.
  3. Therefore all dictators are politicians.

For other types of arguments, informal fallacies are those that can only be detected by examining the content and meaning of the argument:

  1. A poker player is a person.
  2. Therefore a bad poker player is a bad person.

There are many informal fallacies. Some appeal to emotion, others are circular, and still others attempt to discredit the opposing side by attacking it. Different sources classify fallacies differently. Here’s a breakdown of fallacies presented by Hurley in his classic textbook.

Fallacies of Relevance:

A fallacy of relevance is one where the premises are totally irrelevant to the conclusion.

  1. Appeal to force – a naked attempt to threaten the audience:

    If you vote to unionize, we will have to shut down the Toledo plant.

    In many cases, the threat is disguised or presented by a supposedly neutral third party:

    Now I fully support your right to free speech. But if you publish that piece, the party will very likely ostracize you. As we both know it’s not like the old days anymore. You know what that means right? Why would you put your family through that?

    This technique may seem like relatively rare drama but it is used extensively in the legal system. It is standard in plea-bargain deals between prosecutors and the accused. Paired with this fallacy is its opposite, an appeal to personal benefit, also used extensively by the legal system.

  2. Appeal to pity – an attempt to gain sympathy from the audience:

    If I am convicted, my children won’t have a means of support and will suffer.

    The sanctions against Iran are hurting Iranian children who have nothing to do with Iran’s policies. For their sake, let’s stop the sanctions.

  3. Appeal to the people – Using people’s need for admiration, recognition, acceptance, etc. Used alot in advertising:

    You have shown yourself to be a sophisticated and discerning customer. As a result we’d like to offer you our Rubidium card.

    You, American workers, are strong, self-sufficient, and independent! Therefore, you don’t need or want government help.

    You are a respected scientist and an eminent scholar. Your name on our petition will have alot of influence.

  4. Argument against the Person(aka ad hominem attack) – An personal attack to discredit the arguer rather than the argument:

    Bill Maher is against religion, particularly in the public sphere. But Maher is an arrogant, self-righteous jerk. Nothing he says is worth listening to.

    Vladimir Putin argues that historically Ukraine is part of Russia. But we know Putin is an evil dictator who is bent on recreating a Russian empire. Therefore we can discard any historical argument he makes.

    Climate scientists are a bunch of socialists. Therefore climate change is just a socialist plot.

  5. Accident – when a general rule is cited for a specific case it was not intended to cover:

    It’s illegal to be in public in your underwear. So if there’s a fire in your house at night, please get dressed before exiting.

    Stealing someone else’s money is illegal. But the government does it all the time when it taxes me.

    The right to free speech is constitutionally guaranteed. Therefore I shouldn’t be arrested for inciting a riot.

  6. Straw man – creating a new, distorted version of an original argument, destroying the distorted version, and thereby claiming destruction of the original argument.

    John has argued against prayer in school. John is clearly an atheist. But atheism is what they used to have in the Soviet Union. Atheism leads to the suppression of God and its replacement by an authoritarian state. Is that what we want? Of course not.

    Universal health care is a socialist idea. Socialist countries are behind us economically and tell you which doctors you can see. Why would we want that?

  7. Missing the point – when the premises support one conclusion but another, perhaps subtly related, conclusion is drawn.

    My mom told me to stop talking to Bill because he stole my lunch. I guess we’ll have to learn sign language.

    Social spending is abused by recipients and is a source of corruption for politicians. Our only alternative is to stop all spending on social services.

  8. Red herring – similar to missing the point, an attempt to divert attention by changing the subject.

    A warming climate is melting permafrost in arctic regions. That means the people there will have more arable land!

    Climate change is increasing the frequency and severity of Canadian forest fires. Clearly Canada has to get its citizens to be more careful about starting fires outside.

    John called out his representative on the illegal payments he’s been receiving from the Swiss govt. He used four letter words and yelled at him. He was so rude that it was embarrassing to listen to. I wish people wouldn’t act like that in public settings. Totally uncalled for.

Fallacies of Weak Induction

This category contains premises that support the conclusion so weakly that they cannot be taken seriously. Unlike the fallacies of relevance, they provide some support for the conclusion but not enough.

  1. Appeal to unqualified authority – using an authority that is biased or otherwise unqualified. Often the cited authority is perceived to have some qualification in the subject:

    My chemistry teacher said that nuclear power should be outlawed because it is so dangerous.

    The Petroleum Association thinks climate change is a hoax. Therefore we can stop worrying about climate change.

  2. Appeal to ignorance – a definitive argument made based on the idea that nothing has been proved one way or another.

    They keep changing their minds on which foods are healthy or unhealthy. Clearly no one really knows. So I’m just going to keep eating junk food.

    No one has ever proved that God exists or does not exist. Therefore anyone can say anything they want about God and it is valid.

    A stronger variant on this might be:

    No one has ever proved that God does not exist. Therefore he exists.

    No one has ever proved that God exists. Therefore he does not exist.

    Not all appeals to ignorance are a fallacy, however. Our justice system operates on the idea that a defendant is innocent until proven guilty. Therefore obtaining a “not guilty” verdict amounts to showing that no evidence came to light that would implicate the defendant. This is a safeguard intended to give defendants the benefit of the doubt. It does not logically mean the defendant did not commit the crimes he was on trial for.

  3. Hasty generalization – when the sampling used to obtain an inductive generalization is insufficient.

    Silicon Valley Bank, First Republic, and Signature Bank all collapsed recently. Therefore banks are unsafe places to keep your money and you should stop using them.

    Every time I go out in the cold I get sick. Therefore cold temperatures are making me sick.

  4. False Cause – argument based on an imagined causal connection that probably does not exist.

    Successful executives are paid in excess of a million dollars. Therefore, to ensure that our CEO is successful we should raise his salary to at least a million dollars.

    This fallacy is sometimes described as “correlation does not imply causation”:

    Death by drowning correlates with the sale of ice cream. So let’s start regulating the sale of ice cream.

  5. Slippery slope – when the argument rests on an alleged chain reaction when such is unlikely.

    If we legalize marijuana, then cocaine and heroin will follow and more people will start using them. Soon our children will be addicted to heroin.

    If the court rules to outlaw pornography, it will quickly be followed by outlawing all forms of entertainment.

  6. Weak Analogy – when an analogy is not strong enough to support the conclusions drawn.

    Donald Trump and Joe Biden are similar in many ways. They are both old white men. They are both politicians. They both play golf. Biden likes to bike. So Trump should like biking too.

    AI is inherently dangerous because its development is like that of a human child. As the child grows it becomes completely self-aware and capable of autonomous decision making. Once this happens, you can’t be sure if you’ve created an evil monster or a humble servant of humanity.

Fallacies of Presumption, Ambiguity, and Grammatical Analogy

Fallacies of presumption have premises that themselves contain what they purport to prove. Fallacies of ambiguity contain differences in interpretation of content in the premises or conclusion. Fallacies of grammatical analogy are bad arguments that have the same grammatical structure as good arguments.

  1. Begging the question – when a key premise, probably false, is left out, creating the impression that the remaining premises support the conclusion. This can also manifest as a shaky premise being restated as the conclusion, or a circular argument.

    Leaving out a premise:

    Murder is morally wrong. Therefore, abortion is morally wrong.

    The world we live in displays an amazing degree of organization. Therefore the world was created by an intelligent God.

    Shaky premise restated as conclusion:

    The death penalty is justified for murder because it is legitimate and appropriate that someone be put to death for such a heinous crime.

    Circular argument: Reasoning from a first premise, stated as a conclusion, back to that conclusion.

    Ronald Reagan was a good communicator because he spoke effectively.

    You must follow the law because it is illegal to break the law.

    Sam is a great leader because he does great things for the right reasons. He created a high speed train network. He started a program to colonize Mars. He defeated a terrible dictator in war. He did these things because he’s a great leader.

  2. Complex question – when two or more questions are posed as a single question and a single answer is given to them. The answer implies culpability or acquiesence to all the questions:

    Have you stopped lying to the IRS?

    If someone answers “yes” it implies that they used to lie to the IRS and have now stopped. Answering no implies that they are still lying to the IRS. The two obvious questions this conceals are “Have you lied to the IRS?” and “If so, have you stopped?

  3. False dichotomy – the presentation of alternatives as if they are the only ones available and asking for a choice between them.

    Either you buy American or you are not a loyal American. Yesterday you bought a new Nissan. Therefore you are not a loyal American.

  4. Suppressed evidence – not mentioning evidence that could lead to a much different conclusion.

    Herbal supplements are almost always safe. Therefore just go ahead and take the ones you need.

    If it is known that the patient in question has serious allergies to many substances, this would be an example of not mentioning a key piece of evidence.

    A form of suppressed evidence is false equivalence, where we take two pieces of evidence to be equivalent and draw an incorrect conclusion:

    There is no consensus on the age of the Earth because, on one hand, geologists say it is billions of years old and on the other biblical scholars say it is only thousands of years old.

  5. Equivocation – When a word (or phrase) is used in two different senses and an improper conclusion is drawn as a result.

    The constitution gives us the right to bear arms. Therefore I can carry my AR-15 to class with me everyday.

    Here the right to “bear arms” is taken to literally mean the right to carry my gun around (in public) when the intended meaning is more abstract.

    We have a duty to do what is right. We have a right to consume alcohol. Therefore we have a duty to consume alcohol.

  6. Amphiboly – an ambiguity arising from punctuation or grammar leading to the wrong conclusion.

    Oscar Pistorius shot his girlfriend in the bathroom.

    Interpretation 1: The girlfriend was in the bathroom and Oscar was not. Correct phrasing: Oscar shot his girlfriend, who was in the bathroom.

    Interpretation 2: They were both in the bathroom. Correct phrasing: Oscar shot his girlfriend while they were both in the bathroom.

    Interpretation 3: Oscar was in the bathroom and the girlfriend elsewhere. Correct phrasing: Oscar shot his girlfriend while he was in the bathroom.

    Trump, in his will, is leaving Mar a Lago, Trump tower, and Bedminster golf course to Don Jr., Eric, and Ivanka.

    So, who gets what? Do they all 3 get all three properties to share? Or does each one get a different property? In general it is better to just spell it out whenever confusion like this arises:

    Trump, in his will, is leaving Mar a Lago to Don Jr., Trump tower to Eric, and Bedminster golf course to Ivanka.

  7. Composition – the erroneous conclusion that if the parts have an attribute, then the whole has the same attribute.

    Desirable economic growth results when there is low unemployment and low inflation. Therefore we want low economic growth.

    Sodium and chlorine are both deadly poisons. Salt is made up of sodium and chlorine. Therefore salt is a deadly poison.

  8. Division – the opposite of composition, where a statement about the whole is taken to be the same for the parts.

    The Nixon administration was composed of a bunch of crooks. Therefore Henry, Nixon’s White House chef, was a crook.

    The Federation of American Scientists is prestigious and wise. Therefore John, who is a member, is prestigious and wise.

This list of fallacies covers a lot of ground and is very general. But it is by no means the only list. Another, more detailed list can be found here

Some real world arguments and their fallacies

Let’s take a look at some arguments used in the real world and see if we can identify their logical fallacies. In many cases, an argument will have several at once. Some of these are my own and others are taken from Hurley’s textbook.

Parent to child: You have to eat your food. There are alot of starving children in Bangladesh. Child’s response: If you’re worried about starving kids, just send my food to them.

Analysis: The parent uses an appeal to pity. The child uses this to set up a red herring.

After reading “Homosexuals in the Churches”, I’d like to point out that I don’t know any serious, capable exegetes who stumble over Saint Paul’s denunciation of homosexuality. Only a fool (and there seem to be more and more these days) can fail to understand the plain words of Romans, Chapter one. God did not make anyone “gay”. Paul tells us in Romans I that homosexuals become that way because of their own lusts.

Analysis: Appeal to unqualified authority. Ad hominem attack. Arguing in a circle (people are gay because they choose to feel gay (ie lust)).

The notion of buying American is as misguided as the notion of buying Wisconsin. For the same reason that Wisconsin increases its std. of living by trading with the rest of the nation, America increases its std. of living by trading with the world.

Analysis: This argument misses the point. Even if Wisconsin, and the nation, increases its std of living through trade with the outside, the point of buying American is to buy from other Americans who will employ Americans, etc. Buying American is therefore better for Americans than buying from outside America, even if doing either increases our standard of living. The argument is also somewhat of a straw man since setting up a local area as a point for comparison is not really analogous to the whole country. It also suppresses evidence by not acknowledging that trade regulations within the US are quite different than between the US and the world.

The 2nd Amendment guarantees the right to keep and bear arms. But if the founders had been able to see modern society and how we use guns today, they would have been shocked. They would not have written the 2nd Amendment.

Analysis. This argument is a straw man because it sets up an imagined version of reality and then makes a conclusion based on it. It also begs the question by leaving out the premises required to justify its “look to the future” standard of interpretation.

If God is omniscient and omnipotent, I can’t possibly have free will. Since I don’t have free will I’m going to stop trying to be a good person.

Analysis: This argument uses suppressed evidence because it fails to imagine how God feels about this. In fact, deity-based religions reject this idea outright (God still requires you to try). If the arguer is trying to claim that his God doesn’t care if you try, he fails to state this and is therefore guilty of begging the question (by leaving out a premise). If the argument is an attempt to claim that an omniscient & omnipotent deity and free will are incompatible, then the argument is a red herring because it doesn’t bother to elaborate and instead distracts us with a convenient conclusion. This style of argument is usually self serving and makes use of premises that have a superficial appeal but don’t hold up on close inspection.

I can prove that you have a soul using the following argument. Premise 1: I have a machine that can decompose you to your atomic elements and then rebuild you perfectly from those elements. It scans you first and a giant computer stores the exact configuration of atoms in your body. Premise 2: Of course, while you’re dematerialized you’ll be dead. Premise 3: But don’t worry, the machine will bring you back to life, perfectly the same as you are now. Premise 4: The machine can also scan you to learn the exact composition and configuration of elements in your body, mine those elements elsewhere, and create a perfect copy of you. Premise 5: If I ask you if you want to try being decomposed and rebuilt, you will say no. Premise 6: If I ask why you refuse, you will say that the machine can create an exact copy of you from a pool of elements that is not really you. In fact, you could see me doing this and in a few minutes you would be talking to the copied you, sitting right here. Since this now introduces a great deal of uncertainty about whether the rebuilt you (from you own atomic elements) would really be you or just a perfect imitation, you don’t want to go through the procedure. You might die and no one would ever know (except you). Conclusion: You believe in dualism, that there is something other than your physical body. For lack of a better word we will call it a soul.

Analysis: This argument (and its many variants) is actually used in debates between dualists and monists (or physicalists). What is going on here? It might be true that most people will refuse to undergo the dematerialization procedure. It might even be true that most people intuit that they have a “soul”. But that doesn’t really prove dualism, it only proves that most people would react with some trepidation at being brought back from the dead using this process. This argument commits the straw man fallacy. It constructs a distorted and subjective version of reality and gets the subject to accept that. It also commits the appeal to unqualified authority fallacy with its implication that most people would refuse the procedure. Just because most people believe something doesn’t mean its true. And it is an appeal to ignorance and false dichotomy. Since we can get the subject to be uncertain about monism, we will conclude that dualism is the result.

A Christian argument for the Morality of Christian Salvation

Let’s now take a look at a realistic debate between two people and try to identify informal fallacies. This involves a Christian (C) trying to explain Christian salvation to a nonbeliever (N). We preface this by noting that the arguments posed by C are not confined to religion, or any religious faith. They are shared by anyone making a subjective argument. We will discuss this more at some length a little later.

N: I like the idea of a good God but I can’t seem to reconcile some issues around the path to salvation. In Christianity, someone can live their whole life committing heinous crimes like murder and rape but then be admitted into heaven based on accepting Christ late in life (ie a deathbed conversion). Whereas someone who lives a basically good life, doesn’t hurt anyone, but never accepts Christ, ends up condemned. How is that fair? How can I believe in a God that does that?

C: Admittedly you have posed quite the dilemma. I’m troubled by it too and faith is never without thought, questioning, and struggle. But we know that there can only be salvation through Christ. He himself told us that in the Bible. No matter how “good” you are, it doesn’t really matter without Christ.

Analysis: The Christian’s initial attempt is a straightforward appeal to the authority of the Bible, and through it, God himself. In this case, it is an appeal to unqualified authority or, more specifically, unverified authority since beyond its own assertions there is no independent corroboration. Appeal to unqualified authority works in exactly this way. It seeks to end an argument by utilizing a supposedly authoritative source, one many people might actually buy into. The argument also misses the point, since the non-believer is asking a precise question about morality and the reply is simply a statement of what the rules are, the very rules being questioned.

N: That only works if you accept the notion of salvation through Christ to begin with. Then it just becomes a rule that you have to accept, like it or not. The issue for me is how to think my way toward belief and a major stumbling block is the apparent immorality of the situation I just posed. I mean, come on, one guy murdered and raped and the other guy’s only fault was not accepting Christ. Even if that’s how it works, it isn’t a path I’m attracted to.

C: The way I see it, the bad person who eventually accepts Christ deserves a chance to repent and be forgiven. This is fundamental to our faith. That no matter how “bad” you’ve been you still get a chance to repent before Christ by accepting him. Don’t you think that’s fair?

N: Well, yea, I suppose I can get behind the idea of forgiving someone if they’ve truly repented and commit to live a good life thereafter. But what if they are just playing the system? How do we know they’ve truly accepted Christ and renounced their former selves?

C: We may never know but God does. You can’t fake it with God because he knows what’s in your heart. If you’ve truly turned your life over to Him, he will know if you meant it sincerely and will judge you accordingly.

N: Ok, I guess that makes sense.

Analysis: In the first part of the response we have a subtle appeal to pity, one that tends to be quite popular in our culture, that if someone truly repents then they deserve forgiveness. The other part commits a version of the complex question fallacy. Note that two questions are being asked as one. C responds by answering the easy one but the real question concerns playing the system. That is, what if someone plans to live a bad life and repent sincerely at the end? In the Christian system, we still have a bad guy that got away with it, sincere though he may have been in the end. C simply fails to address this problem, although we could see how he might do so by saying that God just doesn’t permit playing the system because he knows when the repentance outweighs the wickedness of the premeditated plan (or something like that). This could also be an example of missing the point since C answers without really getting at the point of the question.

C: Ok, so that leaves us with your supposedly “good” guy who never accepts Christ. Remember that even a “good” guy sins in the eyes of Christ. That is the whole point. Only Christ can forgive you because only he lived a perfect life. So by that standard our good guy isn’t that good, is he?

N: Well, he’s certainly better than the murderer/rapist isn’t he? Isn’t good and bad a matter of degree?

C: Perhaps it is to us humans but God doesn’t work that way. His standard is perfection and no person (except Christ) ever attained it or ever will. That is why we need Him to save us.

Analysis: C here sets up a straw man, that of perfection, and then uses that to characterize the good person as bad. There is no attempt to take on the fundamental issue, which is that the good person is really alot better than the bad person, yet suffers a cruel fate.

N: So another problem is the notion of being exposed to the message of Christ and not accepting it. I’m not going to lie, I’ve heard the Christian message so I guess you could say that by not accepting it, I have rejected it, passive though that rejection may be. But what about people who have never heard of Christ or who live in non-Christian societies where the Word simply isn’t as prevalent? Are they judged as harshly too?

C: You can only attain salvation through faith, and that is true for everyone. But God is also fair, and provides the resources we need to make a decision. He furthermore knows what is in your heart and can perfectly judge whether you have rejected him or not. For someone who has not “heard” in the sense that we understand it, we can be sure that God has provided the information and that the individual has been given the opportunity to freely choose, just like we have. Some of these cases might be hard for humans to judge but we have God to do that for us.

Analysis: Once again C invokes an unqualified (unverified) authority and also tucks in an appeal to ignorance: Since we cannot possibly know the subtle information that God has transmitted to people, it is best not to judge or, better yet, to give Christianity the benefit of the doubt.

N: You try to equate the bad guy with the good guy by saying they are all sinners in the eyes of God. The trouble with a morality that works by placing everyone in the same category is that it is no morality at all. It doesn’t help us become better people. It doesn’t give us a way to judge bad behavior, punish it appropriately, and correct it. It doesn’t give us anything to strive for. It sounds like a setup designed to force us into acceptance of God as the only alternative left. What is moral about that?

C: I see your trouble with it. Before I became a Christian I had a tough time with that too. But as I accepted God, I also gradually accepted His view of a perfect morality, which no one attains.

Analysis: C here simply sidesteps the difficult issue of how he came to faith while claiming to be troubled by these same problems. He just asserts that once he was a believer he understood how the morality worked. C also just ignores N’s appeal to our standard understanding of why we have morality and what it is meant to accomplish. C has basically missed the point.

N: I come back to the basic question I posed above. In one case we have a murderer/rapist who accepted God and was saved and in the other we have a decent guy who didn’t accept God and went to hell. No matter how hard you try to equate them by using a perfect standard of morality (your God’s standard), or try to argue that rejection through passive non-acceptance is punishable by hell, you need to first believe in the Christian concept of God. It is precisely these questions that cause me to hesitate in my attempt to believe in such a God.

C: It is important to realize that God knows what is in your heart and will judge you fairly. It might be hard to describe how he does this in human terms. We will always be bound to earthly ways of judging people. We are limited. But God isn’t. And remember, he wants you to choose Him. If you do that, he will give you the peace you are looking for.

Analysis: We can see by now how C argues. God is always invoked first as the answer to all problems. Here C uses a mixture of appeal to ignorance (we humans just can’t understand) and appeal to unqualified (unverified) authority.

This argument is a staple of Christian apologetics. It contains many fallacies in its details but, writ large, C always refers back to God to answer N. The bad guy’s repentance is sincere (and sufficient) if God deems it so. The good guy is really bad because God maintains a binary standard (perfection or sin). God has a rule that you must be saved through belief in Him. People who live outside Christian societies are given signs by God so they can believe too. In any difficult situation, God has provided the answer, whether we see it or not. This argument is basically an appeal to unqualified authority, although a more specific name (appeal to heaven) exists to describe it.

Note that N is not really asking about the existence of God. He is really just trying to square his version of morality with what appears to be a questionable Christian morality. C never really bothers answering this. In some ways his argument just misses the point. Another name for this fallacy is “avoiding the issue”. The argument might conclude as follows:

N: It seems that your whole point of view rests on accepting Christianity first and then understanding the morality of it. It is clear to me that you didn’t really think your way into Christianity. You just took a leap of faith and now it all works. Do you agree?

C: I certainly accepted Christ at a clear point in my life. If that’s what you call a leap of faith, I’m guilty. But I didn’t do it without thought. Perhaps I didn’t go through the thought process you are going through because everyone is different. But I did think about how Jesus lived and his sacrifice for us. To me the evidence for that is compelling. But that too can be argued about and it might seem different to other people. What was most convincing for me was, having accepted Christ, many of my anxieties just melted away. Now I have someone to talk to, an infinite source of wisdom who will guide me through life. I know that my life is to serve Him. This gives me peace. All the people I know have experienced this outcome once they have become Christians. It is truly life changing and I’d urge you to try it. You don’t have to do much or have an ironclad case in your mind first. Just submit to Jesus in prayer and you will experience what I did.

Analysis: This is C’s coup de grace, so to speak. It contains many fallacies but the main one is an appeal to personal benefit (the opposite of appeal to force).

We might note here that this debate says nothing about the truth of Christianity. The Christian, despite his fallacies of argumentation, might be right in the end. The Nonbeliever, who only exposes the fallacies, does nothing to change this.

The Christian’s predicament here is shared by anyone trying to argue for something through testimony and personal experience. If Bill tells us that he saw cars levitate and fly along our local highway, we would have trouble believing him. He would have to show us the flying cars. If a scientist discovers a new phenomenon in physics, but doesn’t have an experiment that can be reproduced by others, he will have a difficult time convincing anyone. This is like the cold fusion experiments that were done in the 80’s (https://hackmd.io/7B1E6OfAQW-ZJ4NYTdWY3Q?both) but couldn’t be replicated by any other lab. It’s possible the original scientists observed cold fusion but since no one else could do their experiment, no one believed them.

One Christian approach to evidence is to suggest that the nonbeliever try Christianity for themselves. Fair enough, and we presume that enough people have done this that the approach is effective. It is hard to argue this point. Religious faith is intertwined with personal experience in complex ways. If someone “tries” Christianity and it works, who are we to say that they were wrong? Nevertheless, this approach is subjective in nature. Religion still lacks an objective methodology by which to prove itself.

This seems like a serious problem for the religious debater but it shouldn’t be. All manner of human experience is personal and subjective. One person’s meditation, diet, or exercise regimen, one they swear by, is not suitable for everyone. Anyone who makes a subjective claim, religious or otherwise, is bound by the same rules of evidence in converting their claim to an objective one.


Software for Informal Logic

Unlike deductive arguments it is difficult to think how software might solve the problem of identifying fallacies in natural language. The basic problem is one of understanding what is “really” meant. All of this suggests AI so let’s see what it can do.

ChatGPT 3.5 answer to the Christian argument

Let’s try the free version of ChatGPT (3.5) to see if it can figure out the fallacies in an abbreviated form of the Christian salvation argument posed above:

Me: Here is a question posed by a non-Christian to a Christian: “how is it moral for God to forgive a bad person who accepts Christ at the end of his life while condemning a basically good person who never accepts Christ?” The Christian answers the question as follows: “everyone deserves a chance at redemption through Christ, no matter how bad they’ve been. And, Christ’s standard of morality is perfection and no one is perfect. That makes the “good” person just another sinner, like we all are. So the “good” person isn’t really that good.” Can you, ChatGPT, identify the logical fallacies in the Christian’s answer?

ChatGPT: The Christian’s response contains a couple of points that might be analyzed in terms of logical consistency and potential fallacies:

  1. False Dichotomy or Oversimplification: The Christian response appears to present a dichotomy between being a “bad person” who accepts Christ and a “good person” who does not. This oversimplification might be seen as a false dichotomy, as it doesn’t account for various shades of moral character and different reasons for accepting or rejecting Christ.
  2. No True Scotsman Fallacy: The response involves a redefinition of the “good” person by suggesting that, according to Christ’s standard of morality, everyone falls short and is a sinner. This could be seen as a variation of the “No True Scotsman” fallacy, where the Christian is redefining what it means to be “good” in light of their specific theological perspective.
  3. Circular Reasoning: The argument assumes the validity of the Christian theological framework, stating that everyone deserves a chance at redemption through Christ. If the person questioning the morality of this system doesn’t share the same theological beliefs, this response may not be convincing.
  4. Appeal to Consequences: The response implies that everyone deserves a chance at redemption through Christ, perhaps suggesting that the consequences of not having this chance would be morally undesirable. This could be seen as an appeal to consequences rather than addressing the core ethical concern.
    • Analysis:**ChatGPT does a mediocre job of analyzing the argument. First, although it uses a list of fallacies from a different source, this is quite common and not a fault of ChatGPT. The fallacies it comes up with are accepted fallacies. The problmes lie in its reasoning. It correctly identifies the False Dichotomy fallacy in Christian theology, but it misunderstands its role in this argument since both parties have accepted that Christianity works this way. The argument is about the consequences of that acceptance, not the fact of the dichotomy itself. ChatGPT further garbles the reasoning when presenting this fallacy. Furthermore, it misidentifies the distinction between the good guy and Christian perfection as the No True Scotsman Fallacy. This fallacy, although a common one (aka moving the goalpost) is not really being committed here because the argument has this distinction built in from the beginning. Everyone knows the good guy is basically good, he’s just not good enough for Christ. This is precisely what the non-Christian is questioning. ChatGPT’s assertion of Circular Reasoning is somewhat convincing but ChatGPT fails to boil it down to its essence (ie I doubt God because how is the path to salvation moral? Because God said so). The explanation it gives sounds ok on its surface but fails to deliver a clear cut dissection. It’s Appeal to Consequences is beside the point and garbled. Christianity certainly uses this fallacy but it isn’t really pertinent to the argument as posed here. Furthermore, its explanation of the fallacy makes little sense. ChatGPT, or at least ChatGPT 3.5, produces what appears to be a passable analysis but is really quite disappointing.

GPT 4.0’s answer to the Christian argument:

Let’s use GPT 4.0 through Bing (a way of getting 4.0 free) to pose the same question. Here is the response:

GPT 4.0: The Christian’s response to the question posed by the non-Christian contains a hasty generalization fallacy. The Christian assumes that all people are sinners and deserve punishment, but this is not necessarily true. The Christian also assumes that accepting Christ is the only way to avoid punishment, but this is a matter of religious belief and not a universally accepted fact. The Christian’s response also contains a false equivalence fallacy. The Christian equates the “bad” person who accepts Christ with the “good” person who does not accept Christ, implying that they are both sinners and equally deserving of punishment. However, this is a subjective judgment and not necessarily true. It is important to note that this is a complex theological issue that has been debated by Christians for centuries. There are many different views on the subject, and no one answer can fully capture the complexity of the issue. However, the Christian might encourage the non-Christian to explore the Bible and seek answers to their questions. I hope this helps!

    • Analysis:**4.0 does no better than 3.5. It identifies first the hasty generalization fallacy because of the Christian assumption that all people are sinners. No doubt this concept is worthy of questioning, but it is tangential to the argument at hand. It then identifies a false equivalence fallacy because it equates a bad person who accepts Christ with a good person who does not. Well, not exactly, and this represents a lack of understanding on GPT 4.0’s part. The argument equates a good person who does not accept Christ with a bad person, not a bad person who accepts Christ. Furthermore false equivalence here is beside the point since, again, although this concept is worthy of its own study, it doesn’t really pertain to the argument here. The question and the Christian argument starts with the facts of Christian theology which is that all people are sinners and that our human distinction between good and bad is flawed in the eyes of God. None of the protagonists question these ideas. It is assumed that Christianity works this way. The question is how can this be moral.

Overall, ChatGPT does a mediocre job of analyzing this argument for logical fallacies. It is impressive in its ability to put together coherent sounding language and identifies some background fallacies in Christian salvation theology. But it fails to grasp the essence of what is being said by the protagonists and the context in which the argument takes place. Identifying the background issues wasn’t really what it was asked to do. It was asked to identify the logical fallacies in how the Christian answered the non-Christian from the perspective of both parties generally agreeing on what Christian theology says. In a sense, ChatGPT missed the point. Despite this, it would seem that continued development in AI will rectify this problem.

IBM Project Debater

Wikipedia – Project Debater

IBM’s “project debater” allows live debates between humans and the computer. Some of these debates can be watched. The software does not appear to be open source but IBM states that “Key Point Analysis and Narrative Generation are also available as software services through the IBM Research Early Access Program”.

Debater uses the idea of a CDC, a context dependent claim, and tries to find these through a Wikipedia search after being programmed with a question (ie do video games increase violence in teenagers? Debater will find many CDC’s pro and con but only a few are useful. This is because many of them have fallacies and do not contribute to the debate. Clearly Debater is able to assess the quality of informal arguments by identifying fallacies.

Debater uses tools similar to those used in argument mining. In fact it has been called an argument miner rather than an argument inventor.

Here is a debate about preschool funding between IBM’s Project Debater and a human debate champion:

Debate about Preschool Funding

Project Debater, overall, was impressive in her (it’s a she) ability to produce cogent and on-point material. She rolled out statistics and studies on the benefits of preschool with ease. She even invoked a moral duty to help the less fortunate. Her real flaw, however, was not being able to penetrate to the core of her human opponent’s argument, which was that tradeoffs need to be made in funding preschools and that it isn’t the best use of money. Her opponent further argued that funding preschools would amount to a subsidy for the middle and upper classes who send their kids to preschool anyway. Debater could have, for example, shown that the required funding is modest and that there are many examples of far less important government programs which could be used to fund a project like this. She could further have shown that most government programs benefit the middle and upper classes anyway and that programs directed at the poor only tend to be politically untenable. Instead she largely stuck to her original points about the benefits of preschool. She certainly didn’t fail at her task but she gets no better than a passing grade.

Carneades

“tools for argument (re)construction, evaluation, mapping and interchange”

Carneades creates a graph of premises to conclusions and allows an audience to participate in accepting or rejecting the premises. Arguments are assigned weights. A pro and con graph is created for each argument and the pro wins if it outweighs the con (and has accepted premises) and vice versa. Carneades is an argument inventor rather than an argument miner. The latest version dates from 2017.

How_Computational_Tools_Can_Help_Rhetoric_and_Informal_Logic_with_Argument_Invention

A paper on Carneades – The Carneades Argumentation Framework

Pierson vs Post famous property rights case, as argued by the NY supreme court, reconstructed in Carneades

Open Argumentation Platform (OAPl)

“The Open Argumentation Platform (OAPl) pronounced “opal” is a suite of argumentation software that includes APIs, libraries, and user interfaces that work together to support a range of argument-oriented computational tasks and associated pipelines.”

OAPl consists of:

Argument Mining

Argument mining software extracts argument structures from natural language text.

ArgMine:

Academic website with projects on argument mining. Has an argument mining platform where users can enter arguments – ArgMine Platform. Portuguese language only.

Webis Group Research:

Contains several search engines for argumentation, including args (argument search), ChatNoir (web search for arguments), IR Anthology (Papers on information retrieval), Netspeak (writing assistance), Picapica (plagiarism detection), TIRA (experiment execution).

More on args:

A web search tool to identify pro and con sides of any argument. Appears to only use www.debate.org as source material, a defunct website.

ArguAna:

Academic site run by the Webis Group

MARGOT:

Mining arguments from Text. You can enter an argument and it will break it down into claims and evidence. It takes a little while for it to process results.

O’Reilly book on argument mining

Argument Mapping

Argument mapping refers to creating diagrams that enable humans to visualize arguments and facilitate their analysis (by humans). Most of this software is manual in nature.

Rationale
Argunet and Argdown (a markdown language for arguments)
Mindmup
ArgumentMapper
Figjam
Argumentful (includes scoring)