Democrat Doug Jones won the Alabama special Senate election, 50 to 48 percent, which in my Washington Examiner column I likened to picking a lock. But even as Jones won, he carried only one of the state’s seven congressional districts. That’s according to the calculations of J. Michael Coleman, elections analyst for Decision Desk HQ, who’s been producing some excellent maps of election results. Commenters on his tweet were quick to conclude that there’s something fishy here: “gerrymandering,” “gerrymandering even when it’s unnecessary, tbh,” “that’s some serious gerrymandering,” “some beautiful cracking and packing.” The implication is that the Republican legislators who drew the lines did something underhanded.

That’s true, but only if you believe that following the prevailing interpretation of the Voting Rights Act is underhanded. That doctrine is that redistricters must maximize the number of “majority minority districts,” which in the case of Alabama means maximizing the number of black-majority districts. Even the Washington Post’s Christopher Ingraham, in a long Wonkblog article, somewhat grudgingly concedes, in the 12th of his 16 paragraphs, that “part of this may be due to the Voting Rights Act.” Part?

Ingraham also treats the creation of such districts, which necessarily concentrate heavily Democratic black voters in one district, thus leaving adjacent districts more Republican, as a new phenomenon. Not exactly. After the 1990 Census, Republicans combined with black Democrats in Florida, for example, to draw two new black-majority districts, making significant gains in other districts even when Democrats had a legislative majority and the governorship. That cycle — more than 25 years ago — also saw the creation of new black-majority districts in Alabama, Georgia, North Carolina, South Carolina, Texas, and Virginia.

Coleman and Ingraham both point to the rather irregular borders of the black-majority 7th District, noting that it combines black-majority western Black Belt counties with a much more populous black-majority part of Birmingham’s Jefferson County. You might make an argument that Alabama should have two black-majority congressional districts — 29 percent of the seven districts — since 26 percent of its population was black in the 2010 Census. You could make the same argument about South Carolina, which gained a seventh congressional district in the reapportionment following that Census and whose black percentage that year was 28 percent. But, as an excellent map accompanying Ingraham’s article shows, that would be very hard, if not impossible, to do, since concentrations of black residents are scattered across the state. Certainly, the shapes of any two such districts would be much more grotesque than those of the current districts.

It’s also not clear that black Democratic politicians would support the creation of two black-majority (or near-black-majority) districts. In South Carolina, 6th District Democratic incumbent James Clyburn seems to have opposed such a plan, which would have made his district less Democratic and potentially even vulnerable to a Republican challenger. Clyburn is not a fringe figure; he’s the House’s assistant Democratic leader, the number three in the party leadership.

Such fears are not unrealistic. If you look at Coleman’s table showing the Trump and Clinton percentages in each district in 2016 — an election in which Alabama voted pretty much as it usually does — you see that the black-majority 7th District voted 70 to 29 percent Clinton, but that the least Republican of the six other districts voted 63 to 35 percent Trump. More regular district lines might make for at least one marginal district — a goal of redistricting critics — but it would be a marginal district which a Democrat could easily lose.

Lessons to be drawn:

(1) The Democrats’ chief complaint against Alabama’s district lines really lies against the Voting Rights Act, or the prevailing interpretation thereof. Do they want to repeal it? If not, they, as the party whose constituencies (blacks especially, but also Hispanics, gentry liberals and university communities) are geographically clustered, will tend to be at a disadvantage in equal-population redistricting.

(2) Even partisan redistricting plans can backfire if enough voters change their minds. Jones’ 50 percent of the votes enabled him to carry the same number of Alabama’s seven congressional districts — one — as Hillary Clinton’s 34 percent enabled her to carry. But Jones got 47 to 49 percent of the two-party votes in four other congressional districts. If his performance came to be the new normal, Republicans in those districts would be very vulnerable to Democrats. That’s unlikely to happen, unless Steve Bannon can unearth and promote Republican candidates with Roy Moore’s degree of unacceptability — something probably beyond even Bannon’s talents. But I have seen many times, from the 1960s cycle in California to the 2000 cycle in Pennsylvania, how partisan redistricting plans have backfired when some substantial number of voters start voting differently from how they did when the district lines were drawn.