Join the conversation; call 1-877-763-3246, leave a comment below the video, or hit me up on Twitter (@PaulAlanClif)
On today’s Tech, No Babel: Troubleshooting techniques: Affirming the consequent
“Affirming the consequent” is a term in the study of logic. It means, basically, inferring one of many possible causes from the effect.
[tweet “Just because it could be the cause, doesn’t mean it is. Learn why here:”]
The problem with doing this in troubleshooting is that you immediately eliminate possible causes, concentrating, instead on one, which may or may not be the cause.
Here’s an example from my own life. My microphone needs a battery to work. It also needs a working XLR cable. When I recently had trouble with the microphone not working, I thought to myself, “If the battery is dying, the microphone won’t work. The microphone doesn’t work. I’ll change out the battery.” That didn’t fix it. Why? It was a possible cause, but it wasn’t the cause. The actual cause was a short in the XLR cable that connected the microphone to the sound board.
Affirming the consequent is like that. You think of a cause and get convinced that it’s the cause. It’s much better to do it the other way around. What are all the possible causes? Can you eliminate any of them based on other evidence. Then, and only then, do you try testing to see which is the cause.
Keep in mind, though, that sometimes testing will fix a problem when you didn’t purposely address the cause. Moving a cable with a short while changing a battery may make it seem like the battery was the cause, when it was the cable.
So, just because one thing seems to fix the problem, that doesn’t mean it did. You have to be vigilant and keep your eyes open for things that don’t fit.
For more tips on avoiding this mistake, watch the video.
For more questions, comments, or snide remarks, comment below or contact me at: Facebook or Twitter
More video can be found on YouTube.com
Also check out my Google+ page
Transcript: We’ve previously discussed different ways to do troubleshooting and I’ve also discussed a couple of traps to avoid and this is yet another one of those. It’s called affirming the consequent. Now you don’t need to remember what it’s called, you just need to remember what it is and that is a mistake that you can make in your troubleshooting that will lead you down a bad path that just won’t help you whatsoever.
Let me give you an example of affirming the consequent. If the power doesn’t work, the gear won’t work. The gear doesn’t work, so therefore the power must not be working. Is it true that if the gear doesn’t work … If the power doesn’t work, the gear won’t work? Absolutely true. If you have a gear that’s run by electricity and there’s no electricity, it won’t so much work so that’s true. Is it also true though that if the gear doesn’t work, we can backwards look at it and say, “Oh, the cause must be that there is no electricity.”
No, actually we can’t and that’s because doesn’t work could have a bunch of causes. It could be, “Yup, electricity works fine, I just didn’t turn it on.” Could be, “Well, the thing is broken.” Could be a million possibilities so when it comes to affirming the consequent, what it is basically having a general conclusion and trying to go back from a general conclusion to a very specific cause. Let me give you another example, this is one that happened to me and it’s happened a few times so I would think that I would learn my lesson so I have three XLR cables that I use for various reasons and one of them failed but I didn’t know that at the time. I just knew that I record a show and go to edit it and there is no audio there.
If I have the microphone on and if the battery is starting to die, therefore, the sound wouldn’t work. The sound wasn’t working, the microphone was on, those two things are true so therefore it must be that the battery was starting to die so I replace the battery. It didn’t fix it at all not even a little bit so that’s one of the situations where you have multiple possible causes but instead you choose a specific cause and decided that’s it.
What I actually found out was these XLR cables for me, here’s my mic, for me unplugging and plugging it back in, I’d get a little short right there at the connector so not a hard fix, the one I’m using right now has a lifetime warranty so it’s even an easier fix. I just take it back to a guitar center which I know they’re not the most ideal audio place but in my town, a place with a lifetime warranty on an XLR cable, I’m going with that just to save myself the trouble of resoldering which I have to do with my two older cables which I’ll use as backups so all that to say, when it comes to troubleshooting, be careful that you don’t go from something that has multiple possible causes to thinking, “Well, this is the cause I suspect so it’s got to be that.”
No, it doesn’t have to be that. There are probably possible causes that you’re unaware of. In the case of my microphone, I have control over the battery, I have control over the cable but it could be the electronics inside that started to fail. It could be that the soundboard that I’m plugged into has started to fail. It could be a million possible things that have caused this not to work. All things being equal … Tend to be the battery or the XLR cable but I don’t know that until I have troubleshooted.
It’s absolutely the case that you should … Anytime you’re coming up with possible solutions, possible causes rather of a failure, you should try and think of every possible cause and eliminate them one at a time. Is there a reason that that can’t be the case? Going back to our electricity example, if the electricity doesn’t work, the gear won’t work. The gear doesn’t work, is it … Does it make any sense that the electricity isn’t working? Well, you look around and other things are lit up, they’ve got lights, they are normally caused by electricity so that doesn’t seem likely.
You also think, “Wait, the piece of gear that doesn’t work right now, it has lights on it so it’s getting electricity.” Now, that brings it down to is it getting enough or the right kind of electricity or is it something else that’s the cause but you can’t deduce from that the electricity isn’t on altogether because it is. This happened a few years ago where we had an electrical problem in our neighborhood, we don’t often get brownouts so I really hadn’t seen one before but I opened up the refrigerator door and the light in the refrigerator was dim.
I turned on the lights in other parts of the house and the lights were dim and then I noticed that the TV wouldn’t turn on. Other things just wouldn’t work. This was a few years ago when I had a CRT TV so it required more power than what my current television does so you could be having a brownout situation if you’ve got a gear that’s not working so you need to test that. Affirming the consequent comes down to too quickly eliminating other possibilities and just concentrating on one where you have a problem that could have multiple causes and you’ve decided there is one cause.
A classic example is … Well, just imagine this. I have a teenage daughter, I’ve mentioned her before. Her name is actually Trinity and guess where I got the website name. So if my teenage daughter hadn’t eaten anything recently, she would be grumpy, that makes sense. She is grumpy so therefore, she hasn’t eaten anything. That’s affirming the consequent kind of mistake so instead, I can ask her, “Have you eaten anything recently?” She’d say, “Yes.” “Okay, how well did you sleep last night?” Another possible reason that she’s grumpy. “I didn’t sleep well at all. I only slept three hours.”
Okay, now you’ve found something that is more likely to be the cause. It might not be the root cause, we discussed that earlier. It could be that there are other causes that are causing her not to get enough sleep and therefore be grumpy but at least it gets you down the road so thinking about that, when you’re troubleshooting, think, “Okay, I’ve got a general problem, how can I make it more specific? How can I list out all the possible causes that I know? How can I eliminate these possible causes and not just hop on the first possible cause and say well, that clearly has to be it because if that was going on, it would cause the same problem.”
Finally, look for more and more and more evidence to see which possible solution best fits the possible … The actual outcome and I think that will help you in your troubleshooting.