Six points:
1.  The most obvious difference between Vieth and Perry, on the one hand, 
and Larios, on the other, is that the first two are Republican 
gerrymanders, while the last was a Democratic gerrymander.  That would 
explain the point that Hasen made about one difficulty with Pildes's 
analysis:  how to account for the votes of Rehnquist, Thomas, and O'Connor 
in Larios.  (This is not the first time I've asserted partisanship as an 
explanation for O'Connor's votes on gerrymandering cases.  I don't expect 
it to satisfy Pildes or a great many other scholars any better this time 
than it did when I considered it the answer to the blatant contradiction 
between Shaw v. Reno and Quilter v. Voinovich.)

2.  The Larios summary affirmance is another strengthening amendment to the 
full employment act for election law scholars.  Instead of adopting one of 
two bright lines (10% deviation or 0% deviation), the Supremes have allowed 
any point in between, so long as an unspecified justification satisfies one 
or more levels of courts.  This should be a cause of celebration on the list.

3.   Absolute equality in practice may increase, not decrease 
gerrymandering.  Once city, county, and other fixed boundaries must be 
breached, virtually anything becomes possible in redistricting.  Those 
arguing for keeping political or "natural" boundaries automatically lose, 
and almost anything can be justified because of the necessity of avoiding 
population deviations.  While it's true that rough equality limits 
gerrymanders, absolute equality may corrupt absolutely.  That provides some 
rationale for the 10% deviation rule that goes beyond the perhaps more 
obvious one of the census undercount.

4.  The majority of the Supreme Court in Vieth ignored attempts in the 
political science literature, widely used in expert testimony in this 
decade's redistricting cases, to provide manageable standards for judging 
partisan gerrymandering.  Most prominently, Gary King and Andy Gelman 
published a series of articles.  I also published one.  Since Pildes is 
much more knowledgeable in these matters than the majority of the Supreme 
Court (I'm not being snide; he undoubtedly is.), he needs to explain why 
these standards aren't manageable if he's to carry through his argument 
made on the list.

5.  Larios does have bad implications for the Voting Rights Act.  The 
Democratic plan put in multi-seat districts and probably kept some 
inner-city districts that had (undercounted) populations that were too low 
in order to preserve black representation at the same time that it 
represented an effort to keep overall Democratic majorities in both houses 
of the legislature.  (In this sense, it was quite analogous to the ugly 
minority opportunity districts drawn in the 1990s to satisfy both blacks 
and white Democratic incumbents.)  The district court dismissed such aims 
as beneath contempt, or at least consideration.  Combined with the prior 
district and Supreme Court decisions from Georgia in Miller and Abrams, 
during the last decade, this amounts to the following message to 
redistricters:  If you try to create or preserve minority opportunity 
districts (many of these -- see GA v. Ashcroft -- were not 
majority-minority), two things may happen to you -- you'll get slapped down 
for having an unconstitutional purpose, or your purpose will be treated as 
so obviously inconsequential that it won't even be considered in a rational 
basis analysis.  In other words, blacks can only count if they count against.

6.  The remedy in Larios seems even more peculiar now than it did when the 
district court opinion was published.  After the summary affirmance, which 
left Gaffney, etc. standing, a 10% population deviation for state 
legislatures is ok, unless it's biased against one political party.  But 
bias against a political party isn't per se illegal, unless you can figure 
out what satisfies Kennedy.  And since there were no partisan numbers 
produced in connection with the Larios remedial plan or used in concocting 
it, there's no easy to know whether the no-deviation plan was 
unintentionally biased in a partisan or racial  fashion.  Yet there was 
some consideration to such issues, because Persily's initial plan, which 
had a great many black representatives paired, among other things, was 
corrected to eliminate some of those practical consequences of the special 
master's court-enforced ignorance of partisan, incumbent, and racial 
matters.  The best that can be said of this ad hoc solution is that even if 
there are no population deviations, a court-ordered plan can't have a 
too-obvious racial bias, but no one can be given the material with which to 
analyze its partisan and racial bias systematically.  Not only is there no 
manageable standard; no effort to use a manageable standard to assess 
whether a plan is biased is allowed.

In all but the full employment sense, then, I think scholars on this list 
ought to be very disappointed by the great silences of Larios.
Morgan


Prof. of History and Social Science, Caltech
snail mail:  228-77 Caltech, Pasadena, CA 91125
phone 626-395-4080
fax 626-405-9841
home page:  <http://www.hss.caltech.edu/~kousser/Kousser.html> (Newly Revised!)
to order Colorblind Injustice:  http://uncpress.unc.edu/books/T-388.html
          "Peace if possible, Justice at any rate" -- Wendell Phillips