"I know I'm biased, but..." and its equivalents seem to be relatively common in casual conversation--I've encountered the phrase in classroom discussions, on Internet message boards, and in political arguments. In most cases, "I know I'm biased, but..." is used as a way of feigning humility and deflecting criticism by preemptively responding to accusations of bias. That is, the speaker acknowledges that their argument may be flawed in order to deny their opponent the opportunity point out particular biases. It's a way of signaling to the audience, "Yes, there are errors in this line of reasoning, but I already know that, so you can't accuse me of being biased."
But as we all know by now, it's not enough to just acknowledge biases--you have to actually correct the error before you can move on. Admitting that your argument is based on bias does not absolve you of your error, and it doesn't make your argument any truer.
Therefore, "I know I'm biased, but" is a cached thought that we would be better off without. But how can we get rid of it? Tabooing the phrase "I know I'm biased, but..." is not enough, since your brain will probably end up substituting something similar, such as "I may be wrong, but..." instead of making the appropriate correction. Instead, it is necessary to force your brain to consciously think about the bias instead of instinctively rationalizing the biased argument. This is a skill that takes place on the 5-second level: you have to stop your train of thought mid-sentence and think about the situation more clearly. The following should serve as an anti-pattern for when you notice yourself thinking, "I know I'm biased, but...":
1) Stop. I'm not ready to proceed. If there's a bias in my argument, bulldozing over it is never the correct solution. I need to just cut myself off in mid-sentence and think about this.
2) Identify the bias. What is this bias that my brain is trying to cover up? Does it have a name? Where have I read about it before? What heuristic am I using that is causing the problem? Do I have any emotional attachment to this argument that might cloud my judgment? How would I feel if this argument was wrong? Where is my information coming from? Did I do a thorough job researching this argument?
3) Think about potential solutions. What heuristic should I be using instead of the one I am using? Can I substitute a quantitative analysis or Bayesian update instead of jumping to a particular conclusion? Do I need to do more research to determine if this argument is true? What other sources of evidence can I consult?
4) Re-analyze using a different method. What happens when I use the heuristics I just thought about instead of the ones I originally used? What pieces of evidence really support my argument? What facts would need to be different for it to be false? Can I compare multiple perspectives on this argument?
5) Re-evaluate the argument. Does the argument still look correct? Does approaching the problem with a different method yield the same results? Have I completely explained away the bias?
An abstract explanation isn't always enough, so here is an example:
"...and that's why," Albert concluded, "the iPhone is absolutely terrible!"
"I know I'm biased," Barry replied, "but iPhone is the best smartphone on the market!"
Uh-oh, thought Barry. I said that phrase again. Something's not right here. "Hang on a moment..."
Why would I think that the iPhone is the best smartphone on the market? How would I feel if it wasn't the best phone? Well, I'd be kind of annoyed that I spent all that money to buy one. I'd feel disappointed because the advertisement made it look really awesome, and I've always told everyone that it was worth the price. Am I rationalizing this? Hmm, maybe I am rationalizing and I just don't want to believe that I made a bad purchase.
Ok, so what if it is rationalization? What am I supposed to do now? Didn't I read something on LessWrong about this? This feels like "politics is the mind-killer" territory--I should probably be re-thinking my arguments and checking for bias.
But how should I be evaluating the quality of my iPhone? I guess I should ask myself what features I care about--let's pick three. Well, the most important thing to me is service--I make a lot of calls for work and I don't want any of them to be dropped. I want my phone to be durable, too--I'm pretty clumsy and I drop it from time to time. And the phone bill is important too.
Alright, let's add all of this up: The iPhone is pretty fragile, I've already cracked the screen slightly. And it does drop calls sometimes--there might be a network with better coverage, I'm not sure. And the phone bill--my old phone was definitely a lot cheaper, but it also wasn't a smartphone. I'd have to research other networks' coverage and pricing to be sure.
Wow, I might've been wrong about this. That means I wasted a lot of money. And it also means that the iPhone probably isn't "the best" phone out there. Wait, that's not right--it could be the best, but I don't have the evidence to prove it, so my argument isn't right. I have to gather more evidence.
"Are you still there?" Albert frowned in puzzlement. "You kinda fuzzed out there for a second."
"Nevermind," said Barry. "What I should have said was, the iPhone doesn't really do all of the things I want it to do. Say, where's the electronics store?"
Next time you catch yourself thinking, "I know I'm biased, but...", don't let your brain finish the sentence--stop that train of thought and analyze it!
Edit: Many commenters have suggested that "I know I'm biased, but..." is sometimes used to signal being open to counterarguments. As a result, it is best to double-check what you (or your discussion partners) are really signaling so that you can respond appropriately.