However, in considering the policy decisions the Court has made in past few years, the Court is no longer conservative but in fact has swung back across the middle to occupy a position moderately on the left. This process has seen the emergence of OConnor and Kennedy as the all-powerful swing justices who determine the outcome of cases. If either of those two vote with the liberal block of justices, (Souter, Stevens, Breyer, and Ginsberg), the liberal block has the five votes necessary to win any case before the Court.
In the past several years, either OConnor and Kennedy have voted with the liberal minority on just about every significant policy case save one* in which the liberal/conservative justices voted in their respective blocks. In Grutter, OConnor sided with the liberal block in holding that narrowly-tailored affirmative action does not violate the Equal Protection Clause of the Constitution. Then, in Lawrence v. Texas, the Court held that a statute criminalizing homosexual sodomy was unconstitutional. Kennedy wrote the majority opinion in Lawrence, with OConnor filing a concurring opinion. Finally, we have Roper v. Simmons, the death penalty case which came down today. Kennedy was the lone swing justice in this case, writing the opinion holding that the execution of people younger than 18 is cruel and unusual punishment.
Even when the justices do not vote in their respective blocks, the conservative position tends to lose. In Blakely and Booker, the Court has held that judges cannot depart upward in sentencings not based on facts found by a jury and that the federal sentencing guidelines are no longer mandatory. In Hamdi v. Rumsfeld, the Court ruled against the Bush Administration in determining the amount of due process to which an enemy combatant is entitled.
Thus, the myth of the conservative Supreme Court should die today. It wont, due to the ideologues that believe that anything short of the Warren Court is reactionary (just as there are ideologues that believe that anything short of a roll-back to pre-Lochner days is judicial activism), but the analysis is still appropriate. The dominant characterization of this Court as conservative which once applied is no longer illustrative of the decisions which the Court is handing down, and should be discarded.
---I realize that I have likely forgotten cases that I should have included. If so, please indicate them in the comments.
*There may be more than one, but the most prominent one I can think of is Bush v. Gore, admittedly a gigantic exception, but an exception nonetheless.
UPDATE: As I thought, I have missed several cases, some of which do not support my conclusion. For example, in Zelman v. Simmons-Harris, the Court held that the government providing school vouchers to schools with religious affiliations was not unconstitutional so long as the program is one of true private choice and any advancement of religion is only incidental.
Also, in Mitchell v. Helms, the Court held that government distribution of educational materials and equipment both to public and private schools through a school aid program did not violate the Establishment Clause. Thus, in Establishment Clause jurisprudence, the Court has continued to take positions which satisfy conservatives.
The government has lost some high-profile criminal procedure cases. In Kyllo, where the Court held that when the Government uses a device that is not in general public use to explore details of a private home that would previously have been unknowable without physical intrusion, the surveillance is a Fourth Amendment "search," and is presumptively unreasonable without a warrant.
Additionally, in Richards v. Wisconsin, the Court held that the Fourth Amendment does not permit a blanket exception to the knock and announce requirement for felony drug investigations, though it mitigated this by saying that officers may decide not to knock so long as that decision is reasonable (e.g. for the officers' safety).
In Knowles v. Iowa, the Court held that if a police officer stops a driver for a traffic violation, the officer is not permitted to search the car under the rationale that he is performing a search incident to a lawful arrest unless the officer actually arrests the driver.
Moreover, in Dickerson v. United States, the Court declined to overturn Miranda v. Arizona and its progeny.
Still, in recent years the government has won more criminal procedure cases than it has lost. For example, in Ohio v. Robinette, the Court held that the Fourth Amendment does not require that a lawfully seized defendant be advised that he is "free to go" before his consent to search will be recognized as voluntary.
In Texas v. Cobb, the Court held that the Sixth Amendment right to counsel is offense specific, and does not necessarily extend to offenses that are "factually related" to those that have actually been charged.
Ok, I have to go to work now. I know there are cases I have missed (and things I probably have misspelled), so I'll try to update it over lunch.
A very sincere thank you to Eugene Volokh for his suggestions of cases I missed.
UPDATE 2: An emailer adds:
I think that the current court generally sides against the use of government power, whether it be through civil cases that result in damage awards or criminal cases using the police powers. Many of the cases you cited in the con law stuff result in a restriction on police powers or procedural protections for citizens charged with crimes. To those cases I would add a fairly significant strain of cases on statutory matters, which also restrict government power via civil law regulation of conduct:
1. Barnes v. Gorman (2002), which stated that punitive damages are not available in Rehabilitation Act and ADA lawsuits.
2. BMW v. Gore and State Farm (2004), which limit punitive damages to single-digit multipliers in state courts.
3. Great-West Life v. Knudson (2002), which bars "legal" damages for ERISA cases, basically taking fraud and extracontractual remedies off the playing field for employment benefits cases (I disagree with this decision, which was 5-4.)
I think Zelman and Mitchell are two stand-out cases, because they result in a fairly significant expansion of government power to fund religious institutions. Also, Employment Div. v. Smith, which expands the government's ability to prosecute people for religious practices under neutral laws of general application, was a pretty big constitutional turn from the past. That would generally be pro-state police power. These three cases are all of a single piece, I think: allowing government interventions where those interventions are more or less religiously neutral. Whether that is conservative or liberal depends on whether one is talking about civil law or criminal law, generally. An important feature of both types of restriction on government power is that they tend to be activist in nature: the judicial branch is restricting the power of the legislative branch.