Too often, these debates result in defending a default position based on personal preference or emotion more than the serious consideration of the merit of the argument. There are different types of arguments in that defense, and something like the "slippery slope" argument is a perfect example. For example, this is often used to defend the prejudice against gay marriage. The argument that is usually made is that if gays are allowed to marry, what's next-allowing people to marry their animals?
Another popular argument is what I refer to as the "I know a guy" argument. There might be a formal name for it, but basically it is the premise of discounting an argument because someone has first-hand knowledge that, in his or her opinion, disproves the argument.
The classic example is made by smokers in regards to the argument that cigarette smoking is harmful to one's health. Despite overwhelming evidence, someone who smokes and does not want to give the habit up will often rationalize, or counter argue, that their "cousin's uncle smoked three packs a day and lived to be 95 years old."
I am hearing this argument a lot now in the area of vegetarianism/veganism. It seems like many people knew a vegetarian who ran twenty miles a week and died at 34.
Are people lying, or exaggerating? Maybe, but not necessarily. They might indeed know someone that has lived to be very old smoking or eating meat and dairy. Or someone who lived a healthy life and died very young. But maybe, they are hanging onto the hope that what they really know is true, is not. Or, and perhaps most likely, it's a form of justification--providing a reason not to change or a way to sleep at night.
Ultimately, our lives are so personal that I think sometimes we lose perspective and neglect the principles of research and statistics. In the world of statistics, an outlier (or data in the tail of the bell curve) is perfectly normal, even if there are no other factors affecting the result. And whether vegetarians/vegans live longer or smokers die younger, it is not determined on an individual basis, it is based on the normal distribution of that bell curve. Thus if you plotted smokers and non-smokers on a bell curve according to the age of death, the mean score (age of death) of smokers is much less than non-smokers.
On the individual level, and in this example, there are of course a number of factors that can determine how long someone lives. However, knowing someone that differs from the premise being made, does not, in itself defeat the premise. In fact, it's a terrible argument and it's surprising how many people view this as a some sort of trump card.
While I've provided a simple common example, if you listen to conversations--the "I know a guy" argument is more prevalent than it may seem. Although perhaps not as directly related to the bell curve, I heard the argument a lot in the healthcare debate, the "I know a guy from Canada and he had to wait six months to get a surgery he needed." It is also often extended to someone who is personally involved or offended. Consider, for example, the proposal that children of single parents are at a higher risk for drug use than a two parent home. I'll hear, "Well, I'm a single mom and my kids have never done drugs." Or I'll hear it associated with inside personal information, such as "farm animals are not mistreated because I know a farmer and he takes good care of his animals."
The world is a big place, and in the grand scheme of things, we are extraordinarily average. We, and the processes around us, are also very predictable. Though statistics can be difficult to understand, in almost all instances we fall within the normal distribution of a bell curve--falling within one and certainly two standard deviations of the mean.
Unfortunately, even though it may be to some degree counterintuitive, you may indeed know a guy, but it proves nothing.