3. Conclusions are never accepted directly from any source. Conclusions can only be built up using logical rules from accepted data.

This might be the most important tenet of critical thinking. Conclusions are generated from the data they are built upon. Therefore, there is no reason to ever accept any outside conclusions directly into your mind. Instead you can accept the same data, follow the logic rules and, if the data was sound and your process correct, you and the communicating party should have come to the same conclusion.

Let's say you are in an electronics store and the salesperson tells you "This stereo would be perfect for your living room because it's got plenty of power, easy-to-use controls, and an affordable price." The first thing you will do is break down his explicitly-stated data into its component pieces:

• "plenty of power" - Remember context: "compared to what?" You need not be a stereo expert to find out that power for stereos is rated in amps, and then walk around the store comparing the amp output stated for each stereo. This will tell you how it rates compared to other stereos you can buy. Of course, there's also the direct route of testing it out on a pair of speakers in the store and seeing how loud it is, then imaging that in your living room.
• "easy-to-use controls" - You can try out the controls yourself, or go look up some reviews of that model of stereo on the Internet and see what others think of the controls. (The later, although less convenient, will tell you more about the long-term usability of the controls, which you cannot determine in a few minutes of experimentation.) And context: "compared to what?" Try out some other stereos. Maybe they are all easy to use, so this data is essentially worthless.
• "affordable price" - Again, compare to other stereos with the same features and the same amp output. Plus, this has some implicit information: what's affordable for you? That requires data that the saleperson does not have. His conclusion is based on incomplete data.

It would be plain silly for you to accept his conclusion outright, no matter how well-intentioned or apparently reasonable it might seem. But this doesn't just go for salesmen: you never need to accept someone else's conclusion. By getting the data they used, you can reconstruct the conclusion yourself.

This reconstruction process offers several advantages. First, restricting outside entrances to only data means that it is easier to screen out bad information. If a conclusion is good, you should be able to build it from the data. If it isn't, why let it into your head in the first place?

Second, by building the conclusion yourself, you are able to use additional data that was unavailable to the person who communicated it to you. Every person has their own unique perspective and knowledge they can bring to the process. This may result in your getting a different conclusion, which would let you know that either 1. someone's reasoning was faulty, 2. someone doesn't have all the relevant data, or 3. someone has erroneous data. This is a chance to double-check the work, so to speak, and expose errors. It is vastly preferable to expose an error at this point in the process than after a decision has been made, and potential damage has been done.

Once you begin to implement this tenet in your own thinking, you may find it reveals some curious behavior in others. A very common situation is for someone to present an argument with no supporting data (basically just a pre-packaged conclusion) as data. As a critical thinker, you will reject their statement, and instead ask them to provide you with the underlying data. (e.g. "I understand you want me to vote for Smith, but what makes that candidate a better choice than the others?") Sometimes such a request will result in the person responding with either another conclusion ("he'd be the best candidate because he's a real upstanding guy!") or even a flat-out refusal to provide the information ("look, just trust me, it's in your best interest to vote for him"). In this case you should become immediately suspicious of their motives. If they are unwilling to provide you with the data, this probably means one of two things: 1. their conclusion is based on faulty or incomplete data, and/or their reasoning for reaching that conclusion is faulty; or 2. they are telling you a conclusion that they themselves do not necessarily believe, but they wish to influence your behavior a certain way. (The later is common with salespeople - they want you to believe that this stereo is good so that you will buy it and they will get their commission, but they do not necessarily believe it themselves.) Either way, their inability to provide the supporting data so that you may generate the conclusion for yourself is a sign that the conclusion is not worth your time to consider.

So, what are the logical rules by which we build up our conclusions? This is a subject that can (and has) fill volumes, so the treatment here will be light. Most of this process, what we usually call reasoning, is fairly intuitive. Causality, to pick one important aspect of reason, helps us understand when one occurrence is causing another. A few of the rules of causality are:

• An event cannot cause itself, it must be caused by another event or occurrence.
• An event must occur earlier in time than another in order to have caused it. Event A cannot be caused by event B if event B occurred after event A.
• A person or object must be physical proximity to another person or object in order to affect it physically. If Joe is in London he cannot have (directly) caused Sue to trip and fall in Chicago.

However, causality is not the same thing as correlation. For example, I might note that the last four times I've driven to the store in the evening, they've been out of milk. Does this mean that nightfall causes the store to run out of milk? Or that me driving to the store causes them to run out of milk? It seems unlikely. There is a correlation, and that may point us to the true causal relationship; perhaps most people shop in the afternoon after work, causing them to run out; and they do not restock until morning. In this case we say that there is a correlation between the time of day and the availability of milk at the store, but the time of day does not cause the milk to be available or not.

An important aspect of causality is the ordering of events. If we see a wheelbarrow coming around a corner, and shortly thereafter a person holding onto its handles, we might be inclined to believe that the wheelbarrow is pulling the person along behind it. Experimentation, however, might show that separating the person and wheelbarrow would result in the wheelbarrow no longer moving, but the person continuing to do so. Therefore we would conclude that it is the person that was causing the wheelbarrow to move. (Then when a horse followed by a cart comes around the same corner, we might generalize our conclusion to think that the cart is pushing the horse...)

Logic also tends to dip into the realm of the mathematical, using what are know as boolean operations to combine data to produce new data and conclusions. Operations such as AND, OR, and NOT can be used to combine varied pieces of data to produce a final conclusion, which may itself simply be a new piece of data:

Data: Bill is 15 years old.

Data: Bill saw Fredy vs. Jason in the theater.

Data: Freddy vs. Jason is rated R.

Data: To view an an R-rated movie, you must be at least 17 or accompanied by a guardian.

Conclusion: Bill was taken to see the movie by someone, perhaps a parent or sibling.

The art of reason has always been well-regarded in modern society. We have heroes like Sherlock Holmes, whose only claim to fame is their incredible ability to use reason and logic to deduct conclusions based on facts that they collect. I urge the reader to research external sources covering logic and reason more fully. The Wikipedia entries are a good place to start:

http://en.wikipedia.org/wiki/Reason

http://en.wikipedia.org/wiki/Causality