Could you elaborate on this?
It's easiest for me to illustrate through an example. Let's take one specific area of philosophy -- theories of knowledge. For the longest time, philosophers believed there were only 3 necessary and together sufficient conditions for knowledge:
S knows that P if and only if:
1. S believes in P
2. P is true
3. S's belief that P is justified
So for example, let's say I'm S, and P is the proposition that evolution is true. I know that evolution is true if and only if
1. I believe in evolution.
2. Evolution is indeed true
3. I came to believe in evolution through reliable means, such as having read about it in respectable journals and encyclopaedias
The justification condition is really important, otherwise we can just label any belief that happened to be true as knowledge. Let's say I randomly guessed what the lottery numbers were this week -- I still wouldn't to call that knowledge, because I'm not justified. I just got lucky. Note that justification doesn't need to be perfect, or 100% reliable, otherwise we would never know anything. Philosophers were really happy with the JTB (or justified true belief) analysis for a long time.
Then Edmund Gettier came and fucked things up with some counter-examples. Here's a Gettier-styled example:
I'm staring in the meadows and see what seems to be a sheep. I thus come to believe that there is a sheep in the meadow. However, this isn't really a sheep, it's a dog that's disguised as a sheep, and it's very well done. However, as fate would have it, there really is a sheep in the meadow -- one that's hiding behind a rock that I couldn't see. So...
1. I believe that there is a sheep in the meadow
2. There really is a sheep in the meadow
3. I'm justified in believing a sheep in the meadow (afterall, when I usually seem to see a sheep, there really is a sheep)
According to the JTB, I know that there is a sheep in the meadow. But of course, this is highly unintuitive. I just got lucky. So the JTB analysis has been rejected.
So now we try to patch things up, but let me tell you, for every account that's been put out, there's been counter-examples to it. You're stuck in this infinite loop of hypothesis > counter-example > adjusted hypothesis > counter-example > new hypothesis... and on, and on, and on. All the while the Te user is thinking "does all this shit even matter?", "isn't our initial definition good enough?", "are we just showing what a certain term means in English, rather than uncovering some deeper reality?".
Look, I got into philosophy just because I enjoyed it -- I don't mind if our answers aren't all that practical. Having said that, I get impatient when you have to spend an extended period of time grinding over stuff like this. A lot of philosophy is like this, just to varying degrees. I just think a Ti user would be more comfortable seeking this sort of conceptual consistency. I'm thinking of writing about ethics, just so I can remain motivated and grounded.