Last week, I did a press backgrounder on the case, which I set forth below:

Ms. Tapia was convicted of alien smuggling and failure to appear.  When she was captured after her flight from bond, she was found in possession of meth paraphenalia, a sawed-off shotgun and mail belonging to other people, which she later admitted she intended to use to conduct identity theft.  

She was no stranger to the criminal justice system; she had three prior convictions.  Her sentencing range, under the federal sentencing guidelines, was 41-51 months.
 
The district court took all this into account, but also noted that Tapia had been subjected to physical and sexual abuse.   The court observed that the crimes and Tapia's criminal history were "serious," and that it wanted Tapia "to be imprisoned long enough to get the BOP's 500 hour Drug Program."  It accordingly sentenced her to 51 months, the top of the guidelines range.  The Ninth Circuit affirmed, saying that the sentencing statute now in dispute (18 USC 3582) did not prevent the judge from increasing the length of the sentence to promote Tapia's rehabilitation.
 
The Supreme Court took the case to resolve a circuit split on that question.  Two circuits, the Ninth and Eighth, had reached that result, but four others reached the opposite result.  Those four concluded that Section 3582 barred sentencing judges from doing what the judge here did.  They relied on Section 3582's specific language directing that a court, in determining the length of a defendant's prison term, shall do so "recognizing that imprisonment is not an appropriate means of promoting correction and rehabilitation."
 
 As criminal cases go, this one is relatively less remarkable, since it involves, not any issue of constitutional significance, but a relatively obscure statute.  There are, however, some quite interesting features of the case.
 
The first is that the Department of Justice has reversed field, going back on an argument it had made not only quite recently in the Ninth Circuit, but  consistently in administrations of both parties, and had had some success with.  It's extremely unusual to the Department to do that, and even more unusual when the reversal comes so late in the game, after it won with its prior argument in the court immediately below. 

In constitutional litigation, the tradition is that the Department does not reverse course on a prior position unless no reasonable argument can be made in its defense.  That tradition is not as strong in cases of statutory interpretation, but is a salutary standard nonetheless, because it promotes stability, consistency and predictability in the law  --  those being three neutral and very important virtues that largely define law itself.  The fact that a liberal administration follows a more conservative one, or vise versa, should not per se affect the Department's long-held litigating postions, lest ideology and taste supplant stability.  To be fair, the Department's present position is plausible as a matter of statutory construction, and perhaps more persuasive overall that its past postion.  Still, you have to wonder whether that is so to the extent approprate for a mid-litigation reversal.
 
The second noteworthy feature of the case is its casting a spotlight on some unsettled or  --  to be less charitable, incoherent  --  thinking in the law of sentencing.  There is a seemingly schizophrenic divide between those who think the main purpose of sentencing should be to punish and those who think it should be to rehabilitate, and that divide is at the bottom of this case.  The Sentencing Reform Act of 1984 largely rejected rehabilitation as a goal, since the lesson of the late sixties and seventies was that the government was really bad at doing it, if it could be done at all other than by the defendant's own personally motivated change of heart.  But the SRA did not completey reject it, and said, in its main statutory guide, 18 USC 3553(a), that one purpose of sentencing  --  after punishment, incapacitation and deterrence  --  was "to provide the defendant with needed educational or vocational training...or other correctional treatment in the most effective manner." The Court-appointed amicus (which includes Doug Berman as second chair) relies in part on this language to justify the district judge's action.  
 
In my view, the SRA evinced a pretty clear recognition that the main purpose of a sentence was to punish. What we have seeping through this litigation is the confusion that was bound to result when sentencing is viewed, instead, as a struggle about how best to deliver social services to the convict. To do it by lenthening the term of incarceration has the feel of a punitive manuever masquerading as alturism, although I doubt this particular district judge was anything but sincere.  Still, this case probably would not exist if Congress had been clearer about what it actually thought sentencing should or could achieve.
 
The third factor that makes this case interesting and unpredictable is that it follows hard upon a slew of cases in which the Court has been at pains to emphasize that district judges have robust discretion to decide what sentences should be, notwithstanding Congress's views.  This was made quite clear in the Court's 2005 Booker decision, which nullified Congress's explicit direction that the guidelines were to be mandatory, and in the slightly later Kimbrough decision, which held that sentencing judges could point-blank disregard Congress's view of the crack-powder sentencing ratio, and substitute their own policy choices. 

Since then, the Court has continued to emphasize that the most important factor in sentencing is less central control, by either Congress or the Sentencing Commission, than a district judge's individualized discretion.  That almost always works to a defendant's benefit, since "discretion" is virtually always exercised in the direction of shorter prison terms, but this case is an illustration of where it could work the other way.