The Supreme Court has based this rule on institutional competence concerns: circuit court judges are more likely to know and understand state law of local states than are Supreme Court Justices over in Washington, DC. See, e.g., Propper v. Clark, 337 U.S. 472, 486-487 (1949)("In dealing with issues of state law that enter into judgments of federal courts, we are hesitant to overrule decisions by federal courts skilled in the law of particular states unless their conclusions are shown to be unreasonable.").
So far, so good. But now add in Salve Regina College v. Russell, 499 U.S. 22 (1991), a case on whether circuit courts should defer to district court interpretations of state law. In Russell, the Supreme Court held that circuit courts must not defer to district court determinations of state law; review must be de novo. The Supreme Court reasoned that trial judges have little time to give each case, while circuit courts are staffed by legal eagles who are really good at legal research, federal or state. The Court rejected the notion that district court judges from a state are more likely to understand state law than the appellate judges, who usually won't be from that state. The Court found this argument "to be founded fatally on overbroad generalizations":
[T]he bases of state law are as equally communicable to the appellate judges as they are to the district judge. To the extent that the available state law on a controlling issue is so unsettled as to admit of no reasoned divination, we can see no sense in which a district judge's prior exposure or nonexposure to the state judiciary can be said to facilitate the rule of reason.The Court also added a constitutional argument, that the position was foreclosed by Erie Railroad Co. v. Tompkins, 304 U.S. 64 (1938).
So here's the question: Can you reconcile these two lines of cases? If so, how? Why should the Supreme Court defer to Circuit Courts but Circuit Courts not defer to District Courts? (Oh, and no, I'm not writing on this — I'm just curious.)
2) Automatic right to appeal from District Court means Circuit Courts hear far more state law questions than the Supreme Court.
3) Supreme Court jurisdiction only extends to Federal questions, which means state law issues are even rarer.
I'm sure I could think of more if I wasn't ready for a nap.
In neither case is the Court saying that there is anything less than de novo review. But on a question of state law where the Court can't think of a better reason to uphold the Circuit Court, it trots out deference. On the other hand, if it wants to run roughshod over the lower court, it will point out that, at least on unsettled questions, the lower courts are in no better position than anyone else to get it right.
Of course, whenever a lower court gets it wrong, its interpretation has been "shown to be unreasonable." So I don't see a contradiction here at all. Whenever the lower court is acting reasonably, according to settled law, the higher court should defer (because it agrees). But if the higher court disagrees, that makes the law unsettled, and the lower court unreasonable.
In fact, the beauty of Heller is that they don't have to be reconciled at all.
I'm not sure a circuit court interpretation of state law is binding on a district court. Of course, the district judge would rule against the circuit court at its peril. But I would have thought that only state court rulings are binding. Everything else would have persuasive value only, at least as a matter of law.
Supreme court jurisdiction extends wherever the judicial power of the United States extends. This includes alot more than Federal questions. It's true that the Court typically takes cases to resolve Federal questions, but that is for prudential, and not for jurisdictional reasons. If the court decided to take an entire docket of diversity cases involving nothing but state law, it would be a bad decision, but not outside of its power.
Another factor that may be at play is uniformity of interpretation. If Circuit Courts deferred to District Courts on matters of state law, you could wind up with a huge number of different interpretations on that state law in federal court, as each district court judge would be deferred when their opinions went up on appeal. This is a bad thing because we want there to be some level of predictability in the law. On the other hand, this will not occur if the Supreme Court defers to the Circuit Courts on local law, as in most circumstances, any given state's laws are only interpreted by a single Circuit Court.
Since you specifically mentioned the 9th Circuit, why should the Ninth Circuit be more familiar with Montana law than a federal judge in the District of Montana?
Furthermore, I believe that most district court opinions are only persuasive authority on a subsequent district court, whereas most circuit court opinions are mandatory authority on a subsequent appellate panel and may only be overturned en banc.
a. A significant part of the Courts of Appeals' function is error correction, to ensure that the results in individual cases are correct. Fixing erroneous applications of state law would serve that purpose. The Supreme Court, as it sees its role today, does not correct errors. (Leave aside capital cases as an exception that proves the rule.) The Court's job is to clarify and develop federal law. It has no authoritative voice on state law and there is no point in it spending its limited resources on resolving questions of state law which would not be binding on state courts.
b. Both the Supreme Court and the Courts of Appeals also have, as part of their role, ensuring the consistency of the law applied in the federal courts. (This is important whether the law being applied is state or federal.) Generally, review by an individual Court of Appeals will be enough to ensure consistent application of a given state's law, because the majority of cases applying that state's law will go to the Court of Appeals whose territorial jurisdiction encompasses that state. (On preview, this is Dave22's point, or close to it.)
(Perhaps other circuits, if it came to it, should defer to that circuit's interpretation of "its" states' laws? Although the relevant state supreme court would still be able to resolve "splits" of this kind, even without direct review. On preview, this is Armen's point, or close to it. Curse my slow typing.)
The Supreme Court, on the other hand, isn't generally interested in making sure individual cases come out right, but in making sure the circuit courts get the law right. As to federal law, this function justifies their reviewing circuit courts, both to ensure uniformity and because the S. Ct. arguably has a quasi-political legitimacy when it decides difficult legal questions (where "difficult" = untethered to technical sources of legal authority) that the circuit courts don't. But as to state law, there's no reason to think that they have any superiority over the circuit courts, and uniformity is almost never an issue (though I suppose circuits could end up disagreeing over Delaware law or something like that).
Finally, it's worth noting that circuit court judges really can get to know the law of the various states within their circuit. It's not so much a matter of technical expertise, as it is of becoming familiar with the general patterns of a state's law. (State X has a strong policy of consumer protection, or a certain philosophy of water rights, or whatever.)
Judges on the 9th Circuit come from other states than California. The judges in ND Cal, I presume, come from California. It's not too much of a stretch to assume that a judge from California is more familiar with California law than a judge from Arizona or Hawaii or Alaska. Also, its likely that a district judge from N.D. Cal. will have a fairly substantial docket of diversity cases that deal primarily with California law.
2. The gap between the Courts of Appeals and the Supreme Court re: state law issues is much wider than the gap between the Courts of Appeals and the District Courts. In fact, it could be that court of appeal judges see more issues of state law than district judges, just because of the difference in the way appeals and trials work.
3. With the possible exception of the Ninth Circuit, most courts of appeals have few enough states that the judges in each court can gain a good level of familiarity with the laws of each of the states.
Basically, it's hard to imagine that a single district judge is likely to be better at deciding state law questions with his state than three court of appeals judges are at deciding state law questions within their circuit.
Because the supreme court is NOT going to decide contested questions of state law, even in cases where they could do so. They have made it clear they are out of that business.
By contrast, the courts of appeal are obliged to grapple with such questions regularly. There is no reason to suppose the average district judge is better versed in local state law than the average circuit judge. Deferring to nonprecedential district court decisions on questions of law makes little sense, especially where the reasoning seems inadequate or there are conflicting district court decisions- in that case, which would you "defer" to?
I wish I had the wisdom and insight so many of the VC commenters possess. Who knew so many legal questions like this were closed?
The lower courts all claim to be following the US Supreme Court's rulings anyway. Where SCOTUS prefers a change-up, they will foist a change-up. Otherwise they loudly proclaim that the lower courts have spoken.
I realize cases involving state law are often tried in federal courts, and the appeals therefore go to federal appeals courts and SCOTUS. But isn't there a process to certify a question of state law to a state supreme court, instead of leaving the question up to federal judges?
Or, it's easier for one judge to screw up a state law issue than a panel of three or an en banc panel.
The state's supreme court (or whatever it is called in that state) is the final arbiter, but for a variety of reasons, questions of state law wind up in federal court. The most common of these reasons is diversity jurisdiction, where federal courts have jurisdiction over cases involving parties from different states, even if the case is entirely based in state law.
When state law questions are in federal court, it does not have a path to reach any state court through appeals (on occasion a court may certify a question to the state's highest court, but the federal court still has to make the final call). State court decisions are considered to have a precedential value in federal court, but there are often situations where there is no state court decision directly on point. As such, the federal court will have to give it their best shot. The question here is when such a case is in the federal system, which level of the federal judiciary should be deferred to in such a situation.
Okay, I'll bite, because I'm in "fed courts" mode right now (I have the final on Monday): it seems to me that the above bolded language cannot be correct given the Court's holding in Murdock v. City of Memphis (1875). The Court there explicitly said that it only has the appellate jurisdiction given to it by Congress, and that that jurisdiction does not extend to pure state law questions. As far as I am aware, Congress has not changed that since 1875 (although presumably it could). The Court can only decide state law questions to the extent they're antecedent to federal questions.
At least, that's my understanding of things, which could be completely wrong. This stuff is far from intuitive.
Gullyborg, there is a law authorizing the certification of state law questions. I think the question has to be determinative of the litigation or some such thing. But it's rarely used. Some reasons include judges not wanting to burden their colleagues across the street when a bit of research will turn up the answer; state judges gleefully punting difficult questions to their federal colleagues with life tenure; and outright superiority complex of federal judges vis-a-vis state judges.
Also, a circuit-court decision is made by at least two judges, as a rule, so is less likely to be wacked-out than a decision by, say, Judge Kent down Galveston-way.
On one axis, you have the degree of specialization. Lets do this on a scale of 1-10. District courts are most specialized to one state (10 points). The Supreme Court specialized in no state (no points). Courts of appeals are in the middle (lets give them 5 points).
On the other axis, you have the degree of competence. District courts get terrible clerks, at least in the snob view of SCOTUS clerks (no points). The Supreme Court gets great clerks (10 points). Since Supreme Court clerks were once circuit court clerks, courts of appeal get 7 points.
Add this up, and district courts and the Supreme Court both get 10 points, while the Court of Appeals gets 12 points. Thus, we arrive at the conclusion that the Courts of Appeals are the optimal point of deference.
That's just incorrect. The Supreme Court has (with some very minor exceptions) appellate jurisdiction* over everything the Circuits have jurisdiction over, and has appellate jurisdiction over the state courts so long as certain conditions are met. Now, the Supreme Court is required (likely as a Constitutional matter) to defer to state high courts on questions of state law, but they still have jurisdiction over many state law questions and more than the Circuit courts (though the Circuits deal with more state law issues as a practical matter). Indeed, although the Court would likely never do this, it could conceivably grant cert to a state appellate court (assuming there is a basis for federal jurisdiction) after the state high court denied review and then reverse on state law grounds holding that the state intermediate court had misconstrued state law. Again, the odds of this actually happen approach zero, but as a jurisdictional matter it is possible.
*I use "appellate jurisdiction" in a non-technical sense, and include writs of cert, mandamus, etc.
Ryman v. Sears, Roebuck, 2007 WL 2964370 (9th Cir. Oct. 12, 2007).
District Court constructions of state law do not count.
Other circuit courts view the state’s intermediate appellate caselaw as interesting but not dispositive.
Alaska Employment Law
As a recovering lawyer, I do not have access to the appropriate databases to check this, but was any Justice in the majority in all three cases? (adding Buford.) How about Salve Regina College and one of the others? If so, the answer might shed light on Justice's left/right bias versus their theoretical basis for their decisions.
And how to resolve the dilemma? By avoiding its horns. Insist on the certification of state issues to the highest court in the state. (Again, I can't do the research. Is there any state that refuses certification of state issues?) I am not familiar with certification at the District Court level, but some (all?) Circuit Courts have rules governing this.
- The Elk Grove majority was Stevens, Kennedy, Souter, Ginsburg, Breyer, with Scalia recused and O'Connor, Rehnquist, and Thomas concurring in the judgment.
- Regina's majority was Blackmun, Marshall, O'Connor, Scalia, Kennedy, Souter, with Rehnquist, Stevens, and White dissenting.
- Buford was unanimous.
So, Kennedy and Souter were the only two Justices in the majority in all three. O'Connor, Kennedy, Scalia, and Souter were in the majority in Buford and Regina. Not really sure what that tells you, but FWIW.
Some states don't have a certification statute so you can't certify. Also, some states (like California) have Supreme Courts that are notoriously over-burdened (usually with death cases) and thus turn down the certification requests. The California Supreme Court turns down over half of the ones it gets from the 9th Circuit (California won't even permit district judges to certify questions). Also, sometimes, the Circuit will not follow the answer, which actually happened in Idaho I think where the 9th Circuit said thanks but no thanks to an answer it didn't like and the Supreme Court there decided to outright refuse all certifications for a number of years.
If by "smarter" you mean "raised more money for their party," the answer is usually yes.
It is possible for one State to be required to apply the law of a DIFFERENT State. Imagine a car accident in CA when the defendant lives in FL. Yes, personal jurisdiction lies in CA, but no assets do. So sue in FL, and the local court probably will apply CA law. What if the FL court gets it wrong? Should that be expected?
Many certified question statutes do not allow certification from another State. How can this violation of due process be corrected? I think the Supreme Court has the power to grant cert, and the only correct ruling would be taken entirely from State law. Good luck getting cert. granted. This is part of why Justice Stephens favors wide latitude in State choose of law rules. Allstate, 449 U.S. 302, 326 (Stephens, concurring) "Judges are presumably familiar with their own State law and may find it difficult . . . to discover and apply correctly the law of another State." But Justice Stephens probably would agree that FL is required to apply CA law in my hypo.
This ties into a post that appeared here a month or so ago. One Senator complained that a judicial nominee was not from his home state, and thus would lack the proper "jurisprudential" view indigenous to the state. That was a laughable assertion. Why is it any laughable when the Supreme Court seems to be suggesting the same?
What is special insight about California law does Ninth Circuit obtain by being located in Pasadena? Why would someone in Washington, D.C. lack that insight?
I understand deferring to state court judges on interpretation of state law. But I think the rationale for that is more based on federalism concerns than it is the "special insight" living in a state gives one when interpreting laws.
So I think the deference to Circuit Court judges is sort of silly. Deference (whether we're talking levels of review being applied legislation or a lower court's judicial order) seems to be applied in one of three circumstances: 1. out of a genuine concern for separation of powers; 2. as a way to make less work for the Court; and 3. as a way out of tough cases.
Dave 22:
I'm probably arguing a distinction without a difference, because as a practical matter, I agree with you. But here is my argument why your statement isn't quite right. No federal court has the power to declare what state law is. That power only inheres in state courts. Thus, when a Circuit Court renders an opinion about a question of state law, the most that opinion can have is persuasive authority.
If a district court doesn't follow its Circuit on a question of federal law, the reversal language is simple: Case X establishes the law in this Circuit, and the decision of the lower court is contrary to Case X. Therefore, reversed.
If a district court did not follow the circuit on a question of state law, the Circuit would not say anything about the law of the Circuit. And they could not also say authoratatively what the established law of the state was. Instead, the court should say that they had visited the question before and that their best guess of what the State Supreme Court would do is X. That guess has not changed, even in light of the lower court's arguments. Therefore, reversed.
Questions of foreign law fall on something of a sliding scale. Suppose a case hinged on the capitalization requirements of an insurance company under Bahamanian law. It that case, the district court is even empowered, under one of the Federal Rules, to take evidence about the state of the law of Bahama. This kind of finding of foreign law starts to look much more like a finding of fact. And guess what? It looks like District Courts have alot more time to do these arcane undertakings, and higher courts give them alot of deference. Pretty odd, isn't it. How could we let those time pressed District Courts decide arcane points of Bahamanian, or Lithuanian law (two things that came up in my clerkship), but then say they are too incompetent to get some basic American law right.
Anyway, the closer the law comes to home, the more it looks like something that any appellate court could decide.
In reality, the amount of deference that any judge - district or appellate - gets will largely depend on the individual judges involved, and not just on whether they are circuit or district judge. A good district judge, who also served as a state court judge (at any level), will get more deference from a circuit court panel than a judge who plays fast and loose and routinely gets stuff wrong. And the circuit court judges are keenly aware of which is which. Of course, that sort of statement would never see the light of day in a Supreme Court opinion. So instead you get the kind of useless claptrap that masquerades as reasoned rules in the opinions Prof. Kerr cites.
By the way, I think there is some consistency between the two cases. The Court in Salve Regina opted to send the case to the Circuit Court for it to do its de novo review. If Erie really means that all courts have equal access to state law, then I don't see any reason why the Supremes could not have just resolved Salve Regina on its own, and spared the lower courts the burden of a remand.
If I'm right about the result on remand, then all the Salve Regina case means is that there are certain taboo words that the Circuit Court must not utter when performing its incantations over unsettled state law.
What do you mean by Circuit Court opinions are generally made by 2 judges?
Duffy -
I'm now also trying to find out what happened on remand - it went back to 1st circuit appeals court - here's the website to search published opinions. http://www.ca1.uscourts.gov/opinions/main.php I'm not coming up with anything...
I thought this was kind of funny: 501 U.S. 1203
I'm not sure if the opinion is available online, but the official cite is Russell v. Salve Regina College, 938 F.2d 315.
I think he means that it takes two judges to make a majority on the Circuit Court. Nothing mysterious there.
Before I asked about remands I checked the first circuit site and came up with nothing. JonC, thanks for digging up the result.
Now, just think of the number of billable hours that were involved in the preparation for the S.Ct. argument, and then for the remand. What a waste, all so Blackmun could say that the Circuit Court said "abracadabra" when they should have said "presto chango". Does anyone seriously think that this opinion has ever had an impact on the outcome of any case at all, now that we've established that it didn't even have an impact on its own outcome?
1) Should a circuit court defer to district court rulings (within the circuit) on state law (for states within the circuit? E.g., should the Ninth Circuit defer to the N.D. Cal.'s interpretation of California law?
2) Should circuit court X defer to circuit court Y's rulings, or district court rulings from within circuit Y, on state law for states within circuit Y? E.g., should the Second Circuit defer to the Ninth Circuit's (or the N.D. Cal.'s) interpretations of California law?
I think the argument that the answer is "yes" is stronger for the second question than the first. Judges on the Ninth Circuit or the district courts within California see many more cases involving California law, and are more likely to have practiced in front of, or served on the bench in, California state courts, as compared to judges on the Second Circuit who may see one case involving California law once in a blue moon. What do people think?
As for issues of uniformity, the rule would be very sensible if it were to require deference to a District Court in the state whose law is in question, but not otherwise. When splits among judges, or between districts in multi-district states arose, the Appeals Courts could then exercise de novo review in the name of uniformity since there would be no single definite decision to defer to. The issue of District Court rulings with binding value does not seem very problematic as it is a rule of custom, not an absolute binding rule.