One of the important skills tested on law school exams is issue spotting. Although profs differ on how important issue spotting is, I would think virtually all would agree it's an important skill for law students.
Yet we design law school exams to eliminate much of the issue spotting potential. A 1L student prepares for each exam separately, takes each exam separately, and then flushes all the knowledge down the drain to prepare for the next one. A student taking the exam knows that one and only one subject will be tested, and he or she doesn't need to think outside the confines of that subject.
The bar exam isn't like that, however. Both on the multistate and essay portions of the exam, students sit for the exam and read fact patterns without knowing which subject is being tested. The fact pattern could implicate any area of law tested on the exam, or even "crossover" between different subject areas.This means that students need not only to determine which issue within a subject is being tested, but which subjects are being tested, greatly expanding the scope of the analysis required.
It seems there would be a lot of benefit to testing 1L students the same way. Instead of taking a single exam in each individual class, students could sit for a series of exams that could come from any 1L subject or more than one. Because 1L students often take a largely fixed curriculum, it would be possible to administer law school exams in this way. The essays could be divided up according to their predominant subject matter and graded by the relevant professors, with grades assigned as usual.
There would be a number of advantages to conducting 1L exams this way. This approach would encourage students to think about the intersections and boundaries between different subjects. It would give students experience identifying the area(s) of law relevant to a particular fact pattern, not just the issues within a particular area of law. It would encourage students to study for the whole semester's material together, rather than "cramming" for one exam after another. Finally, it would give students early practice with the type of format they will face on the bar exam.
I suppose there would be disadvantages as well. The main one appears to be administrative inconvenience of coordinating and grading exams, especially where "crossover" questions are involved. My suspicion is that students would also dislike the change, even though it wouldn't change the total quantity of work demanded from them. It seems that the advantages would outweigh these disadvantages.