Are the U.S. Sentencing Guidelines Judicial or Legislative?
And Will the Answer Determine Whether They Survive Supreme Court Review?

By VIKRAM DAVID AMAR
Friday, Sep. 17, 2004

In my last column, I began to analyze two cases, United States v. Booker and United States v. Fanfan. In both, the Supreme Court will address the applicability of last Term's blockbuster Blakely v. Washington case to the U.S. Sentencing Guidelines, under which convicted federal criminal defendants are sentenced. (Actually, "Guidelines" is a bit of a misnomer here: The Guidelines are mandatory, binding each federal judge.)

In Blakely, the Court held that under the Sixth Amendment, facts that increase a criminal defendant's possible sentence under a statutory scheme -- that is, that extend the sentence beyond what it could have been absent such facts -- must be presented to a jury, and proven beyond a reasonable doubt. Based on this holding, the Court struck down the State of Washington's sentencing scheme.

Under Blakely, there are a few exceptions: The fact of a prior conviction can be established from public records. And if the defendant concedes a fact -- or waives the right to have the jury determine it -- the fact need not be proven to a jury. That means, in turn, that in a plea agreement, a defendant can give up the right to a jury trial not only on the question of guilt, but also with respect to additional facts that bear on sentencing. (For more detailed background on Blakely, please refer to my last column.)

The federal sentencing system, in some ways, looks a lot like the Washington State sentencing system Blakely invalidated. Does that mean it, too, is a goner? I began to consider that question last time - and will continue the analysis here.

The Federal Sentencing Scheme: In Jeopardy?

Under the Federal Sentencing Guidelines, each federal judge, during sentencing of a given defendant, must choose within a relatively small range of possible prison stays. However, the judge is directed to increase the sentence to a higher range when the judge finds, by a preponderance of the evidence, that certain facts exist. (Examples would include: the fact that the defendant lied under oath at trial, or the fact that the defendant defrauded the victims of a large amount of money.)

Isn't this just what Blakely said a judge couldn't do - increase a sentence based on his own factual findings? The Solicitor General's office says no.

It points out that the scheme struck down in Blakely (and its predecessor Apprendi) created statutory maximum sentences from which judges were given the power to upwardly depart when they found special facts. In contrast, the SG says, the federal sentencing guidelines impose judicial limitations -- not legislative statutory ceilings. After all, the sentencing guidelines are created by an independent judicial body -- the United States Sentencing Commission -- not the legislative branch.

Is the SG correct? It's true that Blakely's (and Apprendi's) language refers to legislative ceilings. In addition, in both cases, it was a legislative rule - that is, a statute -- that limited the upper end of the sentence that would have governed absent the judicial fact-finding. Furthermore, Supreme Court precedents have characterized the U.S. Sentencing Commission - which creates the Guidelines - as an entity within the judicial department, not a legislative body.

But there are some problems, as even the SG's brief acknowledges, with characterizing the Guidelines as completely non-legislative.

The U.S. Sentencing Guidelines: Truly Non-Legislative?

Let's look at the key precedent: Mistretta v. United States. There, the petitioner argued that Congress had impermissibly delegated legislative authority to the Sentencing Commission, thereby violating the constitutional separation of powers.

The Court, however, disagreed. It observed that the Guidelines created by the Commission do "not bind or regulate the primary conduct of the public," as legislative directives typically do. It also noted and that the Commission is not engaged in "the legislative business of determining what conduct should be criminalized" - so its decisionmaking is not like that of a legislature writing the definition section of a criminal statute.

But the Court did note certain ways in which the Guidelines do seem legislative - or at least quasi-legislative. It conceded that, like many federal statutes, the Guidelines were "intended to have substantive effects on public behavior" - here, by deterring criminal activity.

The Court also noted in Mistretta that for 180 days after a given part of the Guidelines is promulgated by the Commission, Congress has the power to modify or disapprove any Guidelines with which it disagrees - and that Congress has done so not infrequently. That leads to a key question: Aren't Guidelines that a legislature can approve or modify at least somewhat legislative in nature?

How Important Is the Legislative/Judicial Distinction Here Anyway?

But there's another, deeper question here, too: Does the legislative/judicial question really matter here?

The Blakely Court gave the defendant the benefit of the lesser sentence the judge would have been constrained to impose, in the absence of additional facts found. In doing so, though, the Court did not really stress where the constraint impose that lesser sentence came from - whether a judicial or legislative source.

Specifically, the Blakely Court described the baseline sentence as the "maximum [the judge] may impose without any additional findings." It could, instead, have described the baseline sentence, as the "most the judge could impose under the statutory scheme without any additional findings." (In some passages, the Court does use the word "statutory," but does not explain the significance of that adjective to the analysis.)

Similarly, the Court said that the facts that must be submitted to a jury rather than a judge are those that "pertain to whether the defendant has a legal right to a lesser sentence." It could have said "statutory right" - but it said "legal right" instead. And the Guidelines have the force of law.

Thus, while Blakely doesn't require that its principles be applied to the Guidelines, it seems to leave open the possibility of that result. But which result is right? To answer that question, we need to look at the constitutional values being promoted by the Blakely/Apprendi line of cases.

What Blakely and Apprendi Permit - and Prohibit

It turns out, though, that figuring out these values is not a simple task. To ascertain them, it's necessary to look at what the Court believes to be permissible - and impermissible - in sentencing.

First, what is permissible? For one thing, it's fine for the legislature to give each sentencing judge pure discretion to sentence within a broad (say, ten-year) range. Apprendi makes plain that judges can "exercise discretion - taking into consideration various factors relating to both offense and offender - in imposing a judgment within the range prescribed by statute."

Inevitably, this discretion may be based on the judge's perception of the facts in the case - but not on formal fact findings. And the facts will be those the judge sees as important - not the ones the legislature has listed.

Moreover, it's also fine, according to the Court, for the legislature to allow judges, when choosing among sentences, to increase a sentence beyond what it would otherwise be when the upward adjustment is based on jury findings of fact beyond a reasonable doubt.

Second, what is impermissible? It's impermissible for the legislature to authorize judges to increase sentences based on their own fact findings.

The Constitutional Values the Distinctions Arguably Serve

Why make these particular distinctions between what is permitted, and what is not permitted? Put another way, what values underlie these distinctions?

One value might be wide judicial sentencing discretion - these distinctions preserve it, by allowing, at least, implicit judicial fact-finding. But if the Court really wants to promote sentencing discretion, it should strike down the federal sentencing guidelines. In the pre-Guidelines world, judges had much more discretion; the idea of the Guidelines was to constrain that discretion somewhat, in the interest of fairness: of treating like cases alike.

Another value might be distrust of the legislature - and in particular, of its ability to choose facts that can bump sentencing ranges high. (My U.C. colleague Aaron Rappaport is working on just such a theory, a preliminary version of which can be seen on Doug Berman's Sentencing Law and Policy Blog.) On this view, it's okay if a judge's view of the facts guides how she exercises her discretion to choose a sentence. But it's not okay if the legislature tells her what facts she has to pay attention to, in particular.

If that is our value, then the legislative/judicial distinction might be thought to be very important - for while the legislature may not be trustworthy to say which facts should guide a judge's discretion, perhaps the independent Sentencing Commission may better deserve our trust. But then we must ask: if distrust of the legislature is key, why is it permissible for the legislature to identify the relevant sentencing facts when those facts are found by juries rather than judges?

This question focuses our attention on the language in Blakely that relates to distrust not of the legislature, but rather, of judges: The Court points out that the Framers wanted juries to be included as part of the "judicial department" so that the "people" would maintain "ultimate . . . control in the judiciary." Just as the Senate acts as a check on the House, the framers intended that juries act as a check on judges.

If this value - of preserving the People's input - is crucial, then it might make sense not to put too much stock in the legislative/judicial (or here, Congress/Sentencing Commission) distinction. Instead, to honor this value, we might simply want to give juries fact-finding power whenever a fact has been publicly, formally identified as important, regardless of who has identified it as such.

Why not, though, also give juries fact-finding power whenever a fact is privately, informally deemed important by the sentencing judge as is the case in an indeterminate sentencing system where a judge simply chooses within a wide range? The answer may simply be a practical one: Such a system would ask the judge to reveal every single fact she thinks is important, and submit each to a jury. But those facts might be numerous - deriving from her own experience in similar cases, formed by long years on the bench, as well as by her observation of witness testimony in a jury trial.

If Jury Involvement Is Truly the Key Value, The Guidelines May Be Doomed

Suppose this third value -- jury involvement - is really the key to what is driving the Court. If so, then the Sentencing Guidelines may be hard to justify - at least insofar as it is not completely impractical to involve juries once a Commission has identified the relevant sentencing factors.

If the Guidelines, as they exist, are partially unconstitutional, will some part of them survive? In legal parlance, the question is one of "severability": If part of a statute violates the Constitution, can the rest survive? Courts answer this question by considering whether the legislature would have wanted the partial statute that results - if given the choice between that, and nothing at all.

Here, there is a way to allow some part of the Guidelines to survive: We could maintain the substance of the Guideline limits, but replace formal, public judge factfinding with formal, public jury factfinding as to all the facts the Guidelines make relevant. The SG, argues, however, that this burdensome, tedious outcome is not what Congress wanted when it authorized to Commission to create the Guidelines.

So, according to the SG, the Guidelines - if unconstitutional -- should simply disappear, and each district judge should be free to sentence within the broad statutory minimum and maximum boundaries for each crime. (The SG argues that in this event, judges can still consider the Guideline ranges for whatever wisdom and concept of equality they reflect, even though such ranges would not be binding on any judge.)

It is hard to know how the Court will react to the severability question. I would say this, however: Suppose, as the SG maintains, it is impractical to have juries make all the factual findings the guidelines identify as important - and Congress would not have wanted this solution. Perhaps it would be practical enough - and in conformity with Congress's intentions - if trial courts could at least advise each jury, prior to its rendering its verdict, that a verdict of guilt could, depending on factual findings that the judge then might make under the Guidelines, lead to a particular sentence range under the guidelines scheme.

Maybe this, then, is a compromise position upon which the Court could settle: Keep the Guidelines, but involve the jury by informing the jury of the potential consequences under the Guidelines of a jury verdict. Perhaps that would do justice to what I take to be among the most important language in Blakely, where the Court said that it was "carrying out the design" of the founders with respect to juries "by ensuring that the judge's authority to sentence derives wholly from the jury's verdict."


Vikram David Amar is a professor of law at the University of California, Hastings College of Law in San Francisco. He is a 1988 graduate of the Yale Law School, and a former clerk to Justice Harry Blackmun. He is a co-author of the Cohen and Varat constitutional law casebook, and a co-author of several volumes of the Wright & Miller treatise on federal practice and procedure. Before teaching, Professor Amar spent a few years at the firm of Gibson, Dunn & Crutcher.

Ads by FindLaw