Passing a test proves you can pass a test, not do the thing on the test.
---
Edit, if you don't like this comment, please consider applying Bayesian logic similar to medical tests. You get both false positives and false negatives.
but what if the thing on the test isn't even the ultimate thing you're trying to do either? there's gotta be some way to assess competence by proxy, no?
And who evaluates the mentors of these apprentices? What if they are the Weinstein of legal practice? Or they have a tattoo of a bowling ball covering up the swastika they had done when they were 17? Or maybe it's the apprentice who has the tat, and is getting preferential treatment.
Just like utilities written in C, the fact that they have not shown obvious flaws yet is no guarantee those flaws don't exist.
I get where your cynicism around standardized testing comes from, but I don't think the solution is to throw up our hands and just trust any rando with an immense responsibility that can have profoundly negative consequences for their clients if done poorly.
How does one solve an arbitrary law problem? Algorithmically. Test questions have solutions, sure. They are guaranteed to have solutions, in linear time. You sit down and write the test in constant time, either you find the solutions, meaning your algorithm is good, or you don't, and have to modify said algorithm.
This modification costs time, and money. It's expensive.
They don't predict the ability to solve any law problem. How could they? How could they guarantee such an algorithmic solution? To an arbitrary law problem?
Is the solution to do away with such testing because of this potential undecidability? No, it is not. Write the bar exam, and pass it. Then move on to law problems that may or may not be solvable.
Comments
Passing a test proves you can pass a test, not do the thing on the test.
---
Edit, if you don't like this comment, please consider applying Bayesian logic similar to medical tests. You get both false positives and false negatives.
but what if the thing on the test isn't even the ultimate thing you're trying to do either? there's gotta be some way to assess competence by proxy, no?
Apprenticeship.
And who evaluates the mentors of these apprentices? What if they are the Weinstein of legal practice? Or they have a tattoo of a bowling ball covering up the swastika they had done when they were 17? Or maybe it's the apprentice who has the tat, and is getting preferential treatment.
Just like utilities written in C, the fact that they have not shown obvious flaws yet is no guarantee those flaws don't exist.
Nah. The more objectivity, the better.
I get where your cynicism around standardized testing comes from, but I don't think the solution is to throw up our hands and just trust any rando with an immense responsibility that can have profoundly negative consequences for their clients if done poorly.
How does one solve an arbitrary law problem? Algorithmically. Test questions have solutions, sure. They are guaranteed to have solutions, in linear time. You sit down and write the test in constant time, either you find the solutions, meaning your algorithm is good, or you don't, and have to modify said algorithm.
This modification costs time, and money. It's expensive.
They don't predict the ability to solve any law problem. How could they? How could they guarantee such an algorithmic solution? To an arbitrary law problem?
Is the solution to do away with such testing because of this potential undecidability? No, it is not. Write the bar exam, and pass it. Then move on to law problems that may or may not be solvable.