Saturday, March 22, 2014

The Ninth Circuit's Minimal Opinion in Haskell v. Harris

Two days ago, eleven judges of the Ninth Circuit issued a long awaited opinion in Haskell v. Harris. 1/ The en banc court ruled against the plaintiffs challenging California's DNA-on-arrest law. However, the court did not address their arguments and has not ended the case one way or the other.

Instead, the court simply said that the Supreme Court's opinion last June in Maryland v. King 2/ means that at least some applications of the California law are constitutional; consequently, "plaintiffs cannot show that the district court abused its discretion in denying a preliminary injunction that would apply to the entire class." The court’s final words were "[i]f plaintiffs believe they're entitled to a preliminary injunction as to a smaller class, they are free to seek it from the district court and we will review it if and when it is presented to us."

The Wrong Class -- or the Wrong Relief?

There is a problem with this logic, at least as the court expressed it. The certified class included “[a]ll persons who are, or will be, compelled to submit to the search and seizure of their body tissue and DNA . . . solely by reason of the fact that they have been arrested for, or charged with, a felony offense by California state or local officials.” As to some members of this class, the en banc court indicated, DNA sampling, analysis, and trawling clearly would be permissible under King. But what could the smaller class be? Who are the people arrested in California who "clearly" are constitutionally subject to DNA sampling and retention as it is done in California?

The en banc court gives no answer, but King most clearly applies when:
The DNA sampling [is] confined to violent crimes and burglaries; officials [have] no discretion to pick and choose which arrestee’s DNA to acquire; a physical intrusion into the body [takes] place; the loci tested reveal[] no sensitive medical or other privacy-laden information; the state trawl[s] the forensic index (of unsolved crime scene profiles) only for matches to the arrestee (and not for partial matches that might point primarily to immediate relatives); the profiling and uploading occur[s] after formal charges; and the samples are destroyed if a conviction does not ensue. 3/
The people in these conditions cannot be part of a class entitled to any relief. However, this excludes, well, no one in California. Even people arrested for violent crimes and burglaries (as in Maryland) are at risk for familial searching (although the Haskell plaintiffs did not rely on this concern as a basis for distinguishing King), and they are not entitled to automatic destruction of samples (as are arrestees in Maryland). Moreover, everyone arrested in California for any felony could have a sample analyzed and the profile trawled before any judge agrees with the state that there is at least probable cause to believe that the arrestee has committed a felony.

Given this set of arguably distinguishing factors, it is not so obvious what "smaller class" the court had in mind. Even after King, the class need not change, but the injunction (if one were warranted) would have to be tailored to different subclasses. For example, arrestees charged with violent crimes or burglaries would not be entitled to the blanket order that plaintiffs requested back in 2009. At that time, plaintiffs asked the district court to "enjoin[] Defendants ... from the compulsory search, seizure, collection, analysis, and/or retention of biological samples from any member of the Plaintiff Class ... until and unless he or she is actually convicted of a felony offense, unless such search, seizure, collection, analysis, and/or retention is supported by a warrant issued on probable cause ... ." 4/ But the entire class still might be entitled to some relief. If we were to suppose that no departure from the Maryland law were permissible —which is the most that the California plaintiffs could hope for — even California arrestees charged with Maryland-type felonies would be entitled to an order requiring the state (1) to delay that profiling and uploading until they are bound over for trial, (2) to destroy their DNA samples if they are not convicted, and (3) to refrain from using their profiles in familial searches.

Because plaintiffs sought only a one-size-fits-all injunction, and that size was too big, the district court correctly denied that particular relief. But variable relief would be consistent with plaintiffs' theories of the Fourth Amendment. This, I think, is what the en banc court was getting at in referring to a smaller class. And, if the court reached a consensus solely on this procedural ground, then it is appropriate to allow plaintiffs to amend their complaint in light of King. 5/

A Concurring Opinion

One judge already disagrees with this construction of the per curiam opinion. Judge Milan Smith, who wrote the panel opinion upholding the California law well before Maryland v. King, 6/ read “the per curiam opinion [as holding] that California's DNA collection law is clearly ‘constitutional as applied to anyone arrested for, or charged with, a felony offense by California state or local officials.’” (Emphasis added).

I doubt that this is a correct reading of the per curiam opinion, but I have to say that the other judges could have been clearer about their desire to avoid opining on the limits of King. The other en banc judges wrote that the case turned on the answer to the question, "Is California's DNA collection scheme constitutional as applied to anyone 'arrested for, or charged with, a felony offense by California state or local officials?'" The answer, they claimed, "is clearly yes." So does "anyone" really mean "everyone," as Judge Smith claims, or does it mean "at least one," as I thought when I first read the per curiam opinion? Only the latter interpretation can be reconciled with the court's willingness to countenance a narrower class action.

Judge Smith denied that a complaint on behalf of a narrower class (or, better, a request for a more nuanced injunction) is viable. Surely, he contended, King decided all these fine points in favor of the state. The arguments in his concurring opinion, however, are not uniformly compelling.

Judge Smith wrote that "[t]he majority in King expressly recognized that its decision 'implicates more than the specific Maryland law.'" 7/ But this observation hardly means that every variation of arrestee sampling in existence in 2013 is constitutional under the inherently vague balancing test adopted by the majority. Of the 28 states with arrestee DNA laws, "11 states require an arraignment or judicial determination of probable cause before a sample can be collected or analyzed." 8/ Suppose that the opinion in King had explicitly announced that this provision is essential to a finding of Fourth Amendment reasonableness. That would have contradicted Judge Smith's understanding of the opinion, but the decision still would have "implicate[d] more than the specific Maryland law." It would have pulled the rug out from under 17 states.

Thus, it is a stretch to infer from the King court's recognition of the obvious — that it was deciding a case of national interest — that is also was deciding on the propriety of every existing DNA-on-arrest law. That is not how the Anglo-American system of case-by-case adjudication works.

Neither does the concurrence gain much mileage from "the sky is falling" rhetoric in the King dissent. Judge Smith accepted at face value Justice Scalia's protestation that the result in King could be — indeed, would have to be — extended to reach the most minor of arrests. In oral argument, Judge Smith suggested that Justice Scalia had to be right. After all, said Judge Smith, "he was in the conference, he heard what the majority decided, and that's how he reads it. How can we read it otherwise?" 9/ Yet, it is a standard ploy of dissenting opinions to exaggerate the scope or sweep of the opinion they oppose, and Justices who participated in a conference on a case frequently disagree over the import of the opinions that eventually emerge.

That said, Judge Smith may well be correct in a Legal Realist sense. It is doubtful that the Justices in the majority in King would respond very differently to the other state laws. 10/ Given the phrasing in Justice Kennedy's opinion, the Chief Justice's earlier stay of the Maryland Court of Appeals decision, and the remarks of several Justices at the oral argument in King, it seems likely that those Justices would have no problem using their balancing test to uphold the California law despite all its differences from the Maryland's law. For the Ninth Circuit to have written an opinion striking down parts of California's system of DNA database system would have perilous. Nevertheless, the court of appeals has left the door open to doing so in this four-and-a-half-year-old case.

  1. No. 10–15152, 2014 WL 1063399 (9th Cir. Mar. 20, 2014) (en banc). The 11 judges, selected at random from the full court, are Chief Judge Alex Kozinski and Judges Harry Pregerson, M. Margaret McKeown, Raymond C. Fisher, Ronald M. Gould, Richard A. Paez, Richard C. Tallman, Johnnie B. Rawlinson, Milan D. Smith, Jr., N. Randy Smith and Paul J. Watford.
  2. 133 S.Ct. 1958 (2013).
  3. David H. Kaye, Why So Contrived? The Fourth Amendment Balancing, Per Se Rules, and DNA Databases After Maryland v. King, 104 J. Crim. L. & Criminology (forthcoming 2014).
  4. Class Action Complaint for Declaratory and Injunctive Relief, Haskell v. Brown, Oct. 7, 2009, at 22-23.
  5. Of course, the ACLU, which represents the plaintiffs, might conclude that it has no chance of prevailing on the theory that states cannot relax a single one of the various limitations written into Maryland's law. For instance, it might give up on those people who are charged with violent crimes or burglaries and contend only that individuals arrested for less serious felonies cannot have their DNA analyzed and uploaded before a judge determines that probable cause is present. This certainly would result in a narrower class. But the court's premise that a subclass of arrestees in California are identical to arrestees in Maryland is faulty. Because the California system does not include automatic sample destruction, everyone whose DNA has been collected on arrest in California has been treated differently. The opinion does not state that this is an insignificant difference, but unless that is what it means, it is hard to see why the ACLU must plead for a narrower class.
  6. Haskell v. Harris, 669 F.3d 1049 (9th Cir. 2012). For criticism of Judge Smith’s theory that “the Constitution's plain text” dictates “the ‘totality of the circumstances’ balancing test to determine whether a warrantless search is reasonable,” id. at 1053–54, see David H. Kaye, On the “Considered Analysis” of DNA Collection Before Conviction, 60 UCLA L. Rev. Disc. 104 (2013).
  7. The majority wrote that
    Twenty-eight States and the Federal Government have adopted laws similar to the Maryland Act authorizing the collection of DNA from some or all arrestees. See Brief for State of California et al. as Amici Curiae 4, n. 1 (States Brief) (collecting state statutes). Although those statutes vary in their particulars, such as what charges require a DNA sample, their similarity means that this case implicates more than the specific Maryland law. At issue is a standard, expanding technology already in widespread use throughout the Nation.
    Maryland v. King, 133 S.Ct. 1958, 1968 (2013). The technology of STR typing is standard and in widespread use, as are computer-searchable databases of the resulting profiles, but this does not make every use of the technology constitutionally acceptable. That the majority referred to the charges that trigger DNA collection as a "particular" that varies among states could signal that these Justices did not think it was terribly important, but the remark does not transform a dictum into a holding and does not mean that every variation is so benign as to be "not constitutionally relevant." Haskell, No. 10–15152, 2014 WL 1063399 (9th Cir. Mar. 20, 2014) (Smith, J., concurring).
  8. Nat'l Inst. of Justice, DNA Sample Collection from Arrestees, Dec. 7, 2012,
  9. C-SPAN, Haskell v. Harris Oral Arguments, Dec. 9, 2013, at 6 min. 34 sec.,
  10. Erin Murphy, License, Registration, Cheek Swab: DNA Testing and the Divided Court, 127 Harv. L. Rev. 161 (2013).

Thanks to Hank Greely and Erin Murphy for emails on these latest opinions in Haskell.
Previous postings on Haskell v. Harris
Previous postings on the opinions in Maryland v. King
  • Maryland v. King No. 1: Quick Thoughts, June 3, 2013,
  • Maryland v. King No. 2: Was There a Search?, June 7, 2013,
  • Maryland v. King: Interlude, June 9, 2013,
  • Maryland v. King No. 3: Bertillonage as Precedent, June 12, 2013,
  • Maryland v. King: A Digression on Ellipses, Actual Innocence, and Dr. Mengele, June 13, 2013,
  • Maryland v. King: “Quite a Worldview”, June 15, 2013,
  • Maryland v. King: The Tenth Justice (Stevens) Votes, June 15, 2013,
  • Maryland v. King: When Being Smart and Witty Is Not Enough, Nov. 27, 2013,
  • Maryland v. King: The Dissent's Ten Second Rule, Nov. 29, 2013,
  • Blast Off for King from Florida's Space Belt? Not Quite, Feb. 26, 2014,

No comments:

Post a Comment