Saturday, September 18, 2010

Inference of Example

We have spent the Summer of 2010 walking through Critical Discussions. We started talking about Claims, moved on to Questioning Claims, and just finished Answering Claims with Evidence is the last newsletter.


Now we are completing the last leg of our journey on Critical Discussions; we will talk about how to evaluate evidence with respect to the claim it is supposed to support. That is, we will examine how a speaker “connects” the evidence to the claim. This is called inference, and it is a VERY important part of the discussion.

There are 6 different types of inference (from strongest “connection” to weakest):

 
  1. Example
  2. Cause
  3. Correlation
  4. Analogy
  5. Narrative
  6. Form
I will discuss each of these in turn.
 

EXAMPLE – When making an “Inference of Example”, the speaker makes a claim and then offers evidence that he/she describes as an EXAMPLE that supports the claim. For instance, I might make a claim of fact that say that all people from England have red hair. In order to support the claim I would show you an Englishman with red hair and say “See? There is an Englishman and he has red hair!” Obviously, this inference is fallacious because we all know that there are blonde, brunette, and bald Englishmen as well. But what if I made the claim about a swan? What if I said all swans are white? Or that all Pandas are black and white? Or that most of our customers are happy? How would you test it?

 
There is one characteristic to look for when testing an example; is the example described truly representative of the population it is purported to represent. Here are a few things to examine when presented with an example:

 
• Has the speaker rushed to a hasty generalization? That is, a case in which there are too few examples, and generalizing that to be representative of the whole class. Is the sample random? Do they represent an appropriate diversity given the range of choices within that population?

 
• Has the speaker offered an overwhelming exception? That is, have they said something that is true, but it is true because they have placed an excessive number of restrictions on the underlying population. For instance, "All Americans are useless at foreign languages. Ok, I'll make an exception for those who live in multi-ethnic neighborhoods, have parents who speak a foreign language, are naturally gifted in languages, have lived abroad or who went to a school with a good foreign language program, but the rest are absolutely useless at foreign languages." By restricting the population to just the few that have had no real opportunity to practice foreign language, the speaker has set up an Overwhelming Exception. The example he gives is restricted to the point that it is no longer representative of the underlying population.

 
• Has the speaker used misleading vividness? That is, have they constructed an example that appeals mainly because of a strong effect it has on the listener’s emotions. For instance “Golf is a very dangerous sport. I know a fellow that was golfing and was run over by a golf cart, badly breaking both legs and leaving him crippled and unable to support his poor family. He eventually was divorced and lost his mind!” In this case, the speaker has used strong, vivid terms to make it seem that golf should be considered very dangerous, as dangerous as skydiving. The experience of the unfortunate man he speaks of is not representative of a significant number of golfing experiences and doesn’t really support the definition of golf as a dangerous sport.

 
• Has the speaker used a special pleading? That is, an interpretation of the facts that indicates that the usual logic is somehow inappropriate and this case is somehow special. For instance, one man shoots and kills another. Our speaker says “We admit that this man shot and killed another, but he did so in self-defense.” Self-defense is a situation that we as a society have agreed serves as a special pleading for kidding – the usual logic (killing must be punished) doesn’t apply in cases of self defense. I often hear this when I claim that 15% of our recent shipments were rejected by customers and I hear the production manager’s “special pleadings” for each shipment. (“Bad info from sales; Design problem; Customer changed mind and is using production defects as an excuse”).

 
It is important to assure that the example is representative and that if there are counterexamples (Englishmen with BLACK hair), they are few.

 
Examples are the strongest type of inference because, if they are truly representative of the underlying population, they are proof as opposed to conjecture. They exist in the real world and can be examined and analyzed. This is not true of some inference types which we will examine later.

 

 
Insist on great business results! Go to Pathfinder Communication