Types of Reasoning
I like thinking and I like being correct. As much as I might want to be, I am not always right. Sometimes it is because I have bad information, sometimes because I open my mouth before checking the facts, sometimes I just don’t think properly. Fortunately, there are lots of folks like me out there who will try to keep me thinking correctly as much as possible. It’s one reason I like to debate, you get challenged, and sometimes get smarter as a result.
I wanted to discuss my three favorite types of reasoning I find useful in an argument.
Deductive Reasoning: Powerful but sometimes Impractical
The gold standard of reasoning for most serious debaters is deductive reasoning. Deductive reasoning is about reaching a conclusion that is certainly true or certainly false. At the heart of any deduction is one or more statements of observed fact called a premise and then a conclusion deduced from that premise. For instance…
Premise: All bloggers are clever
Premise: I am a blogger
Conclusion: I must be clever
The idea being that if the two premises are true, the conclusion is certainly true. Of course, one of the central challenges is whether or not the premises are true or not. While it is largely self-evident that “I am a blogger” there is probably good evidence available disproving that “all bloggers are clever.” Of course, the leap from those premises to the conclusion may seem intuitive to us, but our intuition can often lead us astray in ways that we can demonstrate. To really nail down deductive reasoning we have codified the process into what we call Logic. It has its own laws and rules such as fallacies you can study to become a real pro. I’m giving it short service here and honestly, I’m no expert on the subject.
The power of deductive reasoning is you can come to certain conclusions. Its weakness is that you require certain premises. Many of the subjects we attempt to reason about have a lot of uncertain premises. You can use a conditional premise such as: If all bloggers are clever and I am a blogger then I must be clever. While it holds true, it requires us to answer the first “If” if we want to know the answer to whether I am clever or not. Showing that many bloggers are clever isn’t good enough, we need to demonstrate ALL bloggers are clever. That is not very practical given exactly who is a blogger is constantly changing.
Inductive Reasoning: Playing with the Probable
Inductive reasoning is best understood as being about odds and probability. It is used when there are uncertainties you cannot easily overcome. The argument usually makes an observation about a general pattern, and then draws a probable conclusion from it. An example…
Premise: Most people who cross the street, are trying to get to something on the other side.
Observation: That man in the blue hat is trying to cross the street.
Conclusion: There is likely something on the other side of the street he wants to get to.
The conclusion is not certain. Until we ask the man why he is crossing the street, we can’t be certain what the outcome will be. That said, we can make an educated guess that he’s got similar motivations to most other people we encounter. So much in life is uncertain, that much of the time we have no choice but to use systems of thought that don’t lead to certain conclusions.
Humans are not intuitively good at figuring probabilities. Gamblers often fall into irrational thinking. They figure if they sit at the roulette wheel and loose 9 times in a row, then they are “due” for a win. But in roulette like many other situations involving probability, each spin is independent of the last and you are always a little likely to lose than to win each and every time. We also confuse the improbable with the impossible. Finally, just like with deduction, the premise could be inaccurate or in dispute and we may misjudge what the true probability is.
Abductive Reasoning: Finding the Best-Fit Explanation
In Abductive reasoning, you are looking to explain a set of observations. Often you don’t have all the information needed to make a deductive claim, so instead, you are looking for a best-fit explanation that comports with what you do know. An example.
Observations: Tod did not arrive at the office on time, Tod is usually on time, We called Tod at home and his wife said he left for work at the usual time, We checked the traffic report and there was unusually heavy traffic reported on his way to work.
Conclusion: We think Tod is late because he is stuck in traffic.
With Abductive Reasoning you are typically considering alternate explanations as well and deciding which explanation best fits the known facts. It is possible Tod is at a bar drinking, but that would be out of character for him so it is less a good fit for the known facts. He could have been in an accident, but that is an unusual event and traffic is more commonly a problem than a catastrophic crash.
Humans are pretty good at abductive reasoning and we do a lot of our arguments on these grounds. The most important thing to keep in mind is the conclusions are not certain but are simply well-reasoned guesses as to the truth. Often good abductive reasoning is followed up by a test designed to find the truth of the explanation. The scientific method is very much a process of making an abductive hypothesis and then testing them to further ensure they are the best-fit explanation for the observations.
Deductive and inductive reason can both play a role in abductive reasoning helping rule out a fit or support one. Exactly what criteria you use to determine the best fit is part of the challenge of abductive reasoning. A classic example is Occam’s Razor which says the simplest explanation that fits the facts is the best. Of course, one could argue about what is simple and what is not but it is one of many principles to judge an abductive case.
Another weakness of Abductive reasoning is the thoroughness of our observations. The less information we have, the less certain we can be of what is a good fit or not. And like deductive reasoning, we are depending on the accuracy of our observations. If they are in doubt, they are not good for ensuring a reasonable fit. Finally, we are somewhat limited by our imagination in inventing possible explanations to fit the facts. The correct explanation may be something we have yet to consider and as a result, we pick a weaker and wrong explanation that we are more familiar with.
A note on logical fallacies
Most logical fallacies are specific to deductive arguments. For instance, the Ad-Hominem attack is one of the most commonly cited fallacies. It is a form of Genetic Fallacy meaning it is based on the source of the argument. An Ad-Hominem might go like this. “Anyone who fails at first-grade algebra can’t be right about global warming.” From a Deductive standpoint, this argument has no grounds in proving an argument about global warming is wrong. Nothing about first-grade algebra ensures anything about any topic other than their grade in first-grade algebra or their overall GPA.
That said, this may not be true in an inductive or abductive argument. They do not deal with certainty. Inductively someone who exhibits being wrong on other questions may be considered overall unreliable as a source of information and so we can cast doubt on what they say. Of course, one incidence of being wrong may be unconvincing as establishing the overall probability but it is still not an outright fallacy.
Most arguments are not explicitly one type or another, and most people are not familiar with the different types. Thus before leveling a fallacy, it’s good to consider what kind of argument you think the person making it is implying. Words that convey uncertainty such as “probably” and “most likely” are clear indications of inductive reasoning. Conditionals like “I think” or “from what I can see” are good indicators of abductive reasoning. Statements of great certainty like “absolutely” or “beyond a doubt” will tend to indicate a deductive claim. If you are in a formal argument, it may be a good idea to establish the standard of reasoning on which you are making your claim.