We don't know for sure, but look at the first post in this thread. It's hard to tell which professors are more lenient or stricter in terms of grades, or even to what degree of flexibility a professor has to deviate from any potential suggested curve. Again, check out the first post in this thread for a compilation of people's thoughts on the matter.
Because I'm bored, here's a bunch of information about grades. By way of preface, I looked all this up mostly because I like numbers and find doing this sort of research fun, not because I'm a deranged gunner, even though I realize "deranged gunner" is how this sort of thing comes across on TLS.
Feel free to quote this liberally when grades questions come in next week.
THE HISTORY OF HLS GRADES, PRE-REFORM TO NOW
Back in 2009, the Acting Dean of HLS, Howell Jackson, announced a grade distribution for the new grades (apparently carrying out one of the last Elena Kagan reforms). Prior to this time, the grading system was on an A, B, etc., scale. According to the HL Record, the distribution in 1L classes from 1996 to 2000 (and likely well before and well after that) was: 8-11% A, 17-19%A-, 32-34% B+, 29-32% B, 7-8% B-, and 1% each of A+ and C or lower.
The published grade distribution, according to Above the Law and corroborated by the Crimson, was:
In classes with over 30 JD and LLM students enrolled, the recommended distribution of grades is: 37 percent Honors; 55 percent Pass; and 8 percent Low Pass.... Up to two Dean’s Scholar Prizes per class may be awarded in recognition of outstanding work, provided there are more than 30 JD and LLM students in the course following drop/add.
This began during the 2009-10 school year. This means that we're now in the fifth year of the new grades. (Wow, really? That means my 1L year was the third year. No wonder no one knew what the hell any of it meant.)
In the second year (2010-11), under the new Dean Martha Minow, some stealth changes were introduced (at least, that's Above the Law's way of describing it). The grade distribution returned to being unpublished, although there's no reason to believe that it changed substantially, and, judging by informal conversations I've had with professors, it didn't, with the exceptions of DSs and LPs.
Specifically, according to the Crimson, profs were given "increased discretion over the number of Dean’s Scholar Prizes." No one knows exactly what that means, but guesses earlier in this thread have suggested in the vicinity of 3-5 for a class of 80. According to the HL Record, giving LPs also became discretionary (that is, a prof can give zero). And, judging by the "The curve is suggested in all classes with over 30 students" answer in that interview, the curve became a bit more flexible (that is, a prof can give 31 H's in a class of 80, which is over 38%, and no one cares).
Finally, again according to the Crimson, the current grade point system was introduced, along with — I believe — the current system of calculating Latin honors at graduation.
THE CURRENT GRADE DISTRIBUTION (vs. the old one)
Judging by the above, it's something like this (in PERCENTILES, not percent correct or something):
94-99: DS (discretionary; could be 96-99, or whatever)
0-7: LP (discretionary: could be 0-4, or whatever)
The pre-reform grade distribution was, give or take:
0: C or lower
This was interesting to me, because it means, basically, that the new system is a really good excuse for grade inflation. Mark Weber was paraphrased in the Crimson as having equated the old A+ and the new DS, but the new DS is much more common (there are maybe 3-5 of them per 1L class section, as opposed to at most 1 in the old system). An H mostly overlaps with the old A/A- (63-93 for an H vs. 72-98 for an A/A-), but the cutoff is lower, so some old B+'s also become Hs. The LP overlaps with the old B-/C (0-7 [discretionary] for the LP vs. 0-7 for the B-/C), but because it's discretionary, there can be fewer of them than of the old B-/C.
According to the current HLS grading policy, Latin honors are calculated as follows:
For each class, DS = 5, H = 4, P = 3, LP = 2, and F = 0. Calculate a GPA by calculating a weighted average of your grades for the year (weighted by the number of credits). Then average each of the three years of law school.
Thus, imagine a student with 1 DS and 4 Hs in regular 4-credit 1L classes and all Ps in the other 5 classes (including in both semesters of LRW and a 4-credit elective). That comes out to (from best grade to worst, by number of credits): [(4 * 5) + (4 * 4 * 4) + (3 * 4 * 3) + (2 * 2 * 3) ] / 36 = 3.67 for 1L year. If this student then had two more years of grades, you'd repeat the same calculation for each year individually, and then average the three years. So if 2L year gave 3.33 and 3L year gave 3.6, this student would have an overall GPA of 3.53. This is true despite the fact that the years have wildly different numbers of credits; each year is of equal weight regardless.
The top student gets summa, the next 10% get magna, and the next 30% get cum laude. TLS estimates put the cutoffs for cum laude as around 3.5 or so each year, and for magna around 3.9 or so. (These seem slightly low to me; I was told by a prof that magna is around 4.0, give or take.)
WHAT ALL THIS MEANS FOR A 1L WHO JUST GOT FIRST SEMESTER GRADES
Pretty much nothing. Average and median are about 1-2 Hs. If you did much better than that, don't get too cocky. You still have to do that another five times before you can stamp the "magna" on your transcript. If you did much worse than that, don't freak out. You've got five more semesters to figure it out. Talk to professors and see if you can get feedback. Talk to professors this semester and see if they can advise you how to study and what to pay attention to. Talk to 2Ls and 3Ls. Get outlines and hornbooks. Do something different. You'll be fine.
Even for EIP purposes, it's the total on the year that matters, not just one semester, and a great second semester will help to make up for a so-so first semester (and a crappy second semester can screw up whatever good stuff you've done in your first semester). So take it all with a grain of salt. There's quite a way left to go.
If you did really poorly (multiple LPs and no Hs), talk to OCS early, too. Part of what they paid for is making sure that all Harvard students get jobs when they graduate. They can figure out what you ought to do.