Tuesday, March 30, 2010

Argumentation Schemes (Part 1)

I don't know if anybody is interested in this kind of thing, but in case they are the following is based on a set of handouts I once prepared on argumentation schemes. It is based on work by the argumentation theorist Douglas Walton (taken specifically from this book). He has literally written the book on every informal fallacy out there, worth checking out.

The argumentation schemes here are what Walton calls "common presumptive arguments". A presumptive argument is, according to Walton, not based on deductive nor inductive principles. Instead, it is based on defeasible presumptions. They are far more common in argument than we might care to think, so familiarity with them is essential.

Each image provides the abstract form of the argument, an example and a set of critical questions.

1. Argument from the Position to Know

2. Appeal to Expert Opinion

3. Appeal to Popular Opinion 

4. Argument from Analogy

5. Argument from Correlation to Cause

6. Argument from Positive/Negative Consequences

7. The Slippery Slope Argument 


  1. Well, I am, and I see that these are NOT Walton Douglas's argumentation schemes. Could you give a reference and prove me wrong?

  2. I certainly could:

    Walton, D. Fundamentals of Critical Argumentation (Cambridge University Press, 2006), chapter 3.


    He has slightly different and more detailed versions in different books. But those are the ones that appear in that book.

  3. This comment has been removed by the author.

  4. John D., I know this post is pretty old and maybe you are not going to respond, but I'll give it a shot anyway. There are two things bothering me that are in a way related to your post and I thought maybe you can help me answer them.

    First: What is your opinion on the expression "to give someone the benefit of the doubt"? Why shouldn't the chance of someone telling the truth in the abscence of any other evidence be 50%? Isn't a probability higher than this an unjustified leap of faith?

    Second: If a consensus of scientists claim something laymen cannot check, why should the laymen believe they are telling the truth rather than lying (some sort of a conspiracy)? Is it because conspiraces have turned out to be false more often than not? Is it because scientists in general have earned trust due to the success science has had in our society and the more they are the higher the chances of them telling the truth?

    Thank you for your time.

  5. The last question should have been:
    Is it because scientists in general have earned trust due to the success science has had in our society and the more scientists claim something is true, the higher the chances it is true?

  6. (1) I'm not sure I have much of an opinion on the "benefit of the doubt" principle.

    I would view it as a kind of ethical principle, that is directly concerned with how you treat people relative to outcomes. For example, suppose I am a ticket checker at a train station and I'm suspicious of the ticket offered by one of the passengers. I think he might have come across it by nefarious means. But I'm doubtful because I have limited or no evidence. I would give this passenger the benefit of the doubt if I allow him to continue on his journey.

    What kinds of percentages are involved in giving someone the benefit of doubt? I.e. What level of credence must I attach to the respective possibilities? That would depend entirely on my attitude to the harm or risk that would result from not giving someone the benefit of the doubt. In the criminal law, for example, we tend to be very risk averse and so are quite willing to give someone the benefit of the doubt even if the level of credence we attach to their guilt is over 50%.

    (2) There are conditions under which it is reasonable to believe that a consensus of experts is more likely to be true than the opinion of one expert. There is a whole branch of probability theory looking into this. The classic example is the Condorcet Jury Theorem which states, roughly, that if there are two possible answers to a question, then getting a jury of two or more people to vote on the answers and going with the majority opinion is more likely to get you to the truth. This only works if the jurors are independent and have a greater than 50% chance of being right.

    Jury theorems have been expanded in a number of ways since the time of Condorcet. These expansions show that the basic principle (i.e. going with the majority opinion) holds in other conditions as well. So, I would say it is reasonable for a layman to go with the consensus position whenever the conditions set down in the various jury theorems seem to hold.

  7. John,
    I see you know your way with the justice system. Could you please spare a few thoughts on how you imagine the justice system would look like in a society where people lie more often than they tell the truth?
    I have a darn paper to write on this subject and quite frankly I don't even know if such a society can survive and progress. But even if it can, it would be a very strange place: Someone calling for an ambulance would have to say that there is no emergency in order to be persuasive that there is one :)

    Many thanks and best regards!