It is equivalent to a 4-4 with a recusal.

Neither should 4 votes for control when there are 4 votes against. What you have then is NO LAW.

One other point as there has been a lot of confusion in the media and even in the blogs as to what is “controlling” over the past few weeks. A concurring opinion, by itself, is never controlling. When a Justice concurs, it means that he/she has joined the majority opinion. To the extent that the concurring opinion contains a different gloss than the majority, that opinion contains the views of only a single justice, or at least some number less than the majority.

Such an opinion can be analogized to legislative history that helps interpret the text of the statute. It might be useful for some judges in interpreting the statute, but the text itself is what is controlling. Likewise, the majority opinion is controlling, and the concurring opinion can give you a helpful gloss on the majority opinion, but again, the concurring opinion is not controlling.

The answer is entirely different, however, when you have an opinion concurring in the judgment. When that happens, the concurring Justice does not join the majority or plurality opinion. And if there is no opinion that commands five votes, then we look to plurality opinion or the opinion concurring in judgment, whichever is narrowest, to provide the controlling rule. Determining what is “narrowest,” however, is often the hardest part of the Marks problem.

So although Justice Kennedy didn’t join in Part III-B AS A WHOLE, pieces of Part III-B of Chief Justice Roberts’ opinion are holdings of the Court under the Marks rule.

Thanks for your thoughts. I look forward to reading your article.

You’re right that, to the extent that there is absolute and undeniable disagreement between Justice Kennedy and the plurality on a particular point, the Court has given us no guidance on how to treat such a case. Again, I am still wading through the opinions carefully, and have been researching in Justice Pierce Butler’s papers for the last three hours, so I cannot give you a firm answer, but this presents one of the many problems with the Marks rule. I need to read more carefully, but if your characterization is correct, we may not have a controlling opinion in this case on the issues covered in Part III-B. Tom seems to come to a different conclusion, so again I need time to sort all of this out.

Well, my question then, insofar as III-B is concerned, does Kennedy’s opinion actually concur in any part of III-B, or is there just violent disagreement?

We all know that I, II, III-A and III-C have the imprimatur of the Court. The question is, what to do with the material in III-B? If Kennedy’s views of III-B conflict with the plurality’s, it seems silly to say that Kennedy’s views control insofar as we are concerned with the subject matter of that Part. (I suppose I should have made this clear in my earlier post).

Here, some of the limits on the use of race in the decision are shared between Kennedy and the plurality, and thus are a holding (see Parts III-A and III-C).

But Kennedy’s separate thoughts might or might not be controlling under the Marks rule as refined by Grutter.

This is perhaps one of the most difficult rules to apply (and does not always work well), see, e.g., Price Waterhouse v. Hopkins, but the Court said in Marks v. United States that “[w]hen a fragmented Court decides a case and no single rationale explaining the result enjoys the assent of five Justices, ‘the holding of the Court may be viewed as that position taken by those Members who concurred in the judgments on the narrowest grounds.”

That means that, even though Justice Kennedy’s point of view may not have the support of even a single other Justice, his opinion, to the extent it concurs in the judgment and is the narrowest, is controlling for lower courts under the Marks rule. I personally think it is a terrible rule, and is terribly unhelpful in many cases (which is why I am writing an article about it), but it is a longstanding, important and frequently applied rule, as the presence of several fragmented decisions in just the last few days alone demonstrates. And Mark is right that it controlling in the sense that Justice Powell’s opinion in Bakke was controlling.

What if there were four concurrences? would kennedy’s still be “controlling” because in some sense he provided a fifth vote? I would assume not.

It seems weird to me that the views of the 4 are subservient to the views of the 5th, just because the 4 voted together. Don’t we have to establish some commonality between all *five* to establish that a concurring opinion is controlling? If Kennedy concurred in judgement only, and wrote his own opinion pretty much, would be say that trumped the understandings of the other four?

I haven’t read the opinions — just wondering about the possible effect of a concurring opinion that contradicts or is inconsistent with the plurality. i don’t know that kennedy’s concurrence is contradictory/inconsistent with the portion of the plurality which he did not join, but I think it’s an issue worth examining.

Specifically, I doubt that the 4 in the plurarity agree with Kennedy’s view; thus, unless there is in fact some common ground among the 5 regarding Kennedy’s points (or if the Court has in the past espoused Kennedy’s view), I don’t see how his views would be “controlling.” If there is some narrow issue that was agreed upon by all 5, and the pluraity just took it a little further, and Kennedy didn’t go as far, then I could see that narrow issue is controlling; but if the plurarity and Kennedy are diametrically opposed regarding the portion of the opinion that Kennedy did not join, I don’t think Kennedy’s views should control…

Did that make any sense?